Sensitive magnetometry in challenging environments

Kai-Mei C. Fu,1 Geoffrey Z. Iwata,2, 3 Arne Wickenbrock,2, 3 and Dmitry Budker2, 3, 4 1University of Washington, Physics Department and Electrical and Computer Engineering Department, Seattle, WA, 98105, USA 2Helmholtz-Institut, GSI Helmholtzzentrum f¨ur Schwerionenforschung, 55128 Mainz, Germany 3Johannes Gutenberg-Universit¨atMainz, 55128 Mainz, Germany 4Department of Physics, University of California, Berkeley, CA 94720-7300, USA (Dated: October 13, 2020) State-of-the-art magnetic field measurements performed in shielded environments with carefully controlled conditions rarely reflect the realities of those applications envisioned in the introductions of peer-reviewed publications. Nevertheless, significant advances in magnetometer sensitivity have been accompanied by serious attempts to bring these magnetometers into the challenging working environments in which they are often required. This review discusses the ways in which various (predominantly optically-pumped) magnetometer technologies have been adapted for use in a wide range of noisy and physically demanding environments.

I. INTRODUCTION of the sensor with the environment itself - it would be a difficult task indeed to convince a doctor to place a hot rubidium or cesium vapor cell inside a patient’s body, no Magnetic fields are routinely measured with high sen- matter how well protected. sitivity to probe the physical processes that underlie a Unsurprisingly, utilization of enhanced measurement vast array of natural phenomena. From geological move- techniques and noise mitigation is an integral part of ments, solar flares, and atmospheric discharge, to inter- magnetometer development across many applications, cellular processes, neuronal communication in the brain, and is well documented across the literature. Notwith- molecular scale chemical processes, and sub-atomic in- standing, recent advances in sensors motivate new crite- teractions; magnetic fields are produced and can convey ria for evaluating the best match between sensor and ap- detailed information about both the dynamic and static plication, and the techniques to optimize measurements. properties of these systems. While no single magnetometer type is suited for all sens- While the ubiquity and high information density of ing challenges, the diversity in available magnetometer these fields make them attractive to measure, many of technologies can cover the envisioned plethora of envi- the field sources to be studied exist in environments ronments. that are hostile to the state-of-the-art precision mag- Here, we describe how optically pumped magnetome- netometry techniques developed over the last 60 years.√ ters (OPMs) based on hot atomic vapors as well as neg- Magnetometers with sensitivities better than 10 pT/ Hz atively charged nitrogen-vacancy (NV−) defects in dia- must contend with measuring signals in environments mond are particularly well suited for these challenges, with formidable magnetic noise and dramatic gradients. and the strategies that have been developed to adapt Whether in a laboratory, hospital, airport, moving car, these sensors to the broad range of measurement condi- helicopter or spacecraft, the background environmental tions. noise must be understood and surmounted. Neurons fir- ing in the brain have an associated magnetic field sig- nal on the scalp surface lasting a couple ms with a few II. OPTICALLY-PUMPED MAGNETOMETERS hundred fT amplitude (i.e., only parts per billion of the Earth field), meaning movement within even a 1 nT/cm There is a large variety in the ways to measure mag- gradient would swamp the monitored brain signal. netic fields in extreme environments. In this review, we Beyond these difficulties of signal-to-noise, magne- discuss mainly (but not exclusively) two types of devices tometers may also be required to perform under phys- that belong to the same broad class of optically pumped

arXiv:2008.00082v3 [physics.atom-ph] 12 Oct 2020 ical conditions which at first glance may seem incom- magnetometers (OPM): magnetometers based on glass patible with their operating principles. High- and low- cells filled with vapors of alkali atoms and magnetometers pressure and temperature environments affect the per- utilizing color-center defects in solids and more specifi- formance of almost any magnetometer. High levels of cally, the workhorse of the field, nitrogen-vacancy (NV) radiation, whether in space, at particle accelerators or centers in diamond. Both magnetometer types imple- at nuclear reactors, could disrupt the sensitive quantum ment precise measurements of the response of the re- states prepared in magnetometers, or compromise sensor spective spin system to magnetic fields, to infer the local infrastructure and electronics. Vibrations can make cer- magnetic field. Both magnetometer types can be initial- tain measurements impossible, while a rapidly changing ized (“pumped”) to a specific spin state, controlled and field may be far beyond a sensor’s bandwidth and dy- read out with laser light (see Fig. 1). Both types of de- namic range. We must also consider the compatibility vices have received growing attention in recent years and 2

FIG. 1. Optical magnetometry with vapor cells (a-d) and with NV centers (e-h) emphasising the similarities. (a) An example of a glass vapor cell used for magnetometry. It is two centimeters in diameter, has a reservoir for Rb and a stem to separate the sensing volume and the reservoir. The inner walls of the cell are coated with an alkene film [1] enabling coherence times of up to 77 s. (b) Basic working principle (admitting numerous variations). Individual Rb atoms flying with thermal velocities ballistically through the cell are subject to pump laser light to polarize the magnetic moments, a magnetic field B0, an oscil- lating magnetic B1-field to probe the magnetic resonance and an additional probe laser field to read out the spin-precession non-destructively via optical rotation of its polarization. (c) The Rb magnetic-sublevel manifolds of the F = 1 ground and excited states. Optical pumping with a circular polarized on-resonant D1-line laser light at 795 nm transfers the atoms into the mF = +1 sublevel (quantization axis along B0). If the oscillating B1-field corresponds to the energy difference between Zeeman sublevels, there is a coherent transverse component of the magnetization that manifests itself as polarization oscillation at the Larmor frequency ωL in the far detuned linear polarized probe beam. (d) The corresponding magnetic resonance appears after demodulating this oscillating signal with a lock-in amplifier (LIA). Its center frequency is, to first approximation, proportional to the magnetic field. This magnetic-field measurement scheme is referred to as Mz magnetometry [2]. (e) A selection of diamond samples having been exposed to different doses of electron irradiation but not yet annealed. Irradia- tion and annealing are steps in producing samples with different color-center densities. Room-temperature coherence times can be milliseconds. (f) Basic working principle (admitting numerous variations). NV centers are embedded in a diamond matrix and conventionally subjected to green 532 nm light for polarization, a magnetic field B0, and an oscillating magnetic B1-field to probe the magnetic resonance. The measured signal from the NV centers is either spin-state dependent red photoluminescence (PL) or infrared absorption (not shown). (g) The relevant energy levels of the NV centers: the magnetic triplet ground and 3 3 excited states ( A2 and E, respectively) separated by an energy corresponding to the 637 nm zero-phonon line and the two 1 1 singlet levels E and A1. The grey gradients above the energy levels indicate broadening due to phonons. A green 532 nm 3 3 laser is often used to excite the A2 → E transitions. Even though the excitation and subsequent PL are spin-conserving, the intersystem crossing (ISC) to the singlet states is not. The rates are higher for the mS = ±1 excited-state sublevels, which leads to accumulation of NV population in the mS = 0 sublevel. Magnetic resonances can be observed as a reduction of photoluminescence when the frequency of the oscillating magnetic field B1 matches the energy difference between magnetic sublevels. On resonance, population is transferred from the mS = 0 into one of the mS = ±1 sublevels. The mS = ±1 states have a higher energy due to spin-spin interactions. (h) The two spin transitions can be observed around a central frequency of D ≈ 2870 MHz, with the difference of the two center frequencies determined by the magnetic field. Important quantities are the contrast C and the width of the resonance as a measure for the coherence time of the NV center (ensemble). 3 are being applied for measurements in various challeng- (b) 1.00 (c) 10 1.00 ing environments due to their unique properties. 8 0.99 1 6 11 Vapor-cell magnetometry,√ with a√ demonstrated sensi- 9 2 10 −18 0.99 3 9 tivity of ≈ 160 aT/ Hz (10 T/ Hz) by M. Romalis’ 0.98 4 8 group at Princeton [3], is comparable to state-of-the-art 5 7 PL [norm.] 1 7 PL [norm.] 0.98 SQUID systems (superconducting quantum interference 0.97 3 5 devices) which require cryogenics [4, 5]. The sensitiv- 0.96 2 4 6 ity of vapor-cell magnetometers improves with size; for 0.96450 500 45 50 55 60 65 95 100 105 110 a given density the more atoms one can probe the bet- B-Field [mT] B-Field [mT] ter. This scaling is advantageous for spatially extended (a) 1.10 field measurements with the trade-off of limited spatial 1.08 resolution. These sensors usually incorporate glass cells 1.06 filled with gas, which are macroscopic objects with di- mensions typically between 0.2 and 10 cm (Fig. 9c) and 1.04 are not well suited for probing magnetic structures below 1.02 these dimensions. 1.00 NV-NV NV-center magnetometers are typically less sensitive 0.98 Contrast 4.5%

than vapor cells, although continuously improving [6]. Photoluminescence [norm.] NV-P1 0.96 GSLAC This lower-sensitivity is offset by the flexibility provided 0 20 40 60 80 100 120 by the solid-state host matrix. The solid-state sensors B-Field [mT] operate from cryogenic temperatures [7] to over 700 K [8] and from high-vacuum [7] to 60 GPa [9]. Moreover, there FIG. 2. Photoluminescence of a an ensemble of NV centers are sensors based on single NV centers nanometers from illuminated with green light as a function of applied magnetic the diamond surface [10], in nanofabricated scanning field. (a) An “overview” scan over a broad magnetic field probe tips [11, 12], and in diamond nanoparticles (nan- range. (b) Detail around 51.2 mT. Visible are features that odiamond) [13]. This flexibility enables unprecedented can be attributed to cross-relaxation with substitutional ni- small sample-sensor distance, nanometer spatial resolu- trogen atoms, so called P1-centers (1-7). Features 8-10 are due to cross-relaxation with NV centers oriented along other tion and incorporation into biological systems. While the three axis of the diamond crystal. (c) Detail around the solid-state focus of this review is the NV-center in dia- ground-state level anticrossing (GSLAC) where the Zeeman- mond, we note that the field is under rapid development. shifted electronic mS = −1 state is close in energy to the Solid-state magnetometers have also been demonstrated mS = 0 state [25, 26]. (Reprinted from [27] , with permission in SiC, a technologically advanced wide-band gap semi- of AIP Publishing.) conductor [14], and electrical readout of magnetic sensors in both diamond and SiC has been realized [15–18]. In the following sections, we discuss how OPMs can be the effect of microwave fields can be replaced by that of used in various challenging environments including high- modulated light. Various kinds of modulation were stud- noise environments, physically extreme environments ied and applied to magnetometers over the years [21, 22]. and biological environments. An example of the necessity to use all-optical techniques Magnetic sensors employ elements such as heated va- is stand-off and mesospheric magnetometry (Sec. IV G 2). por cells or cooled cryogenic components, as well as mi- It turns out that, in certain cases, not only scalar but crowave and low-frequency magnetic fields. They can also all-optical vector magnetometry is possible [23, 24]. thus influence a nearby sample or other proximal sen- Here the idea is that modulated bias magnetic fields are sors, which could be a problem for dense sensor ar- replaced with effective magnetic fields produced by light rays required for applications like magnetoencephalog- shifts. raphy (Sec V C) or for experiments searching for parity- and time-reversal-invariance-violating electric dipole mo- ments [19]. Radiofrequency fields used in magnetometers B. Microwave-free diamond magnetometry can also exert unwanted influence on the sample and ad- jacent sensors. This problem can be mitigated with the All-optical magnetometry has so far been possible with methods discussed in the following subsections. NV centers either with relatively low sensitivity using the broad dependence of NV fluorescence near zero field (see Fig. 2(a) or at cryogenic temperatures, where NV centers A. All-optical magnetometry exhibit the effect of electromagnetically induced trans- parency (EIT) (see Ref. [28] and references therein). At All-optical magnetometry, where no other fields but higher temperatures, optical transitions of NV centers light are applied to the sensing element, in the case of preserve spin projections and no EIT effects have been atomic-vapor magnetometers goes back to the seminal observed so far. Magnetometry can still be performed work of Bell and Bloom [20] who realized that, for atoms, without applying microwave fields employing a bias dc 4 magnetic field to “tune” near one of the sharp features into a smaller sensor area. The technique is successfully seen in Fig. 2. The sharp dependence of the signal on used in a variety of sensors including Hall probes [33], magnetic field near these features enables magnetometric magnetoresistance sensors [34, 35], magnetic tunnel junc- sensitivities comparable to those obtained with the tech- tions [36], superconducting quantum interferometer de- niques employing microwave fields [27, 29]. Extensions of vices (SQUIDs) [37], atomic-vapor magnetometers [38] the MW-free magnetometry technique include its combi- and, most recently, NV-based magnetometers [39]. nation with cavity-enhanced absorption readout [25] and MW-free vector magnetometry [30]. We note in passing that a technique for microwave-free electric sensing with NV centers was proposed recently [31]. The amplification factor  obtained utilizing flux- concentration techniques can approach three orders of magnitude [33, 40], however there are practical trade- III. THE SIGNAL-TO-NOISE CHALLENGE offs. The sensing area is fundamentally increased, lim- iting the sensor’s spatial resolution. For example, for Magnetometers with high sensitivity are attractive for the flux-concentrator NV magnetometer system (Fig.3a), measuring minute signals, but an experimenter must also the flux concentrator had dimensions ≈30 times or more contend with the environment in which the measure- of those of the diamond sensor [39]. The concentrator- ment will be made. Relevant magnetic noise sources are enhanced sensitivity could be, in principle, alternatively ubiquitous in the laboratory, field, and urban environ- achieved by scaling up the sensor size, but this approach ment. Power-distribution lines produce oscillating mag- may be practically limited by the available light power netic fields with amplitudes 10-20 nT at a distance of 30 and the field-homogeneity requirement. Additional fac- meters. A car passing by, or an elevator moving at a tors that should be taken into account when using mag- distance of 10 meters produces fields of a few nT. Even netic flux concentrators include the filtering of off-axis away from man-made sources of noise, one needs to con- fields, nonlinearity and saturation, remnant fields and tend with natural noise√ of the geomagnetic field which sensitivity to temperature via ferrite thermal noise. A is typically at the 1 pT/ Hz level at 1 Hz (with approx- detailed analysis taking into account thermal eddy cur- imately 1/f increase towards lower frequencies) as well rent and magnetization noise showed that for low fre- as gradients on the order of 20 pT/m or even larger [32]. quencies and low-conductivity ferrite materials, magne- This necessarily creates challenges for measuring signals tization noise dominates [39]. Magnetization noise scales at the fT level. In addition to multitudes of transient linearly as the square root of the magnetic loss factor noise sources, real-world environments may also contain µ00/µ02
in which µ = µ0 − iµ00 is the complex mag- √ background fields that are large in magnitude or have netic permeability. Most recently, 195 fT/ Hz sensitiv- large gradients. These features can create serious techni- ity has been achieved with a flux concentrator utilizing a cal challenges for magnetometers with limited dynamic low-noise permalloy and a 472 magnification factor [41]. range, can be additional sources of noise and can com- Further significant improvements will require the identi- promise magnetometer performance. fication of lower-noise materials. Despite these disadvan- The general approach to addressing the “signal-to- tages, there are several reasons flux concentrators may be noise” challenge consists in a combination of measures to desirable in practical devices. Most field applications do enhance the signal and reduce the noise. In this section not require fT sensitivity and the use of a concentrator re- we review the common techniques utilized in the magne- laxes power, size, and weight requirements. We note that tometry community to isolate sub-nT level magnetic sig- a second novel application of flux concentrators is signal nals from the noisy environment. Many techniques are modulation (Sec. III C). The signal is modulated via the engineering tools common to all magnetic-sensor plat- magnetic permeability by utilizing a small coil wrapped forms, however for concreteness we include examples in around the ferrite concentrator [42] or mechanically mov- specific platforms in which each technique has been suc- ing the concentrator with a micro-electromechanical sys- cessfully implemented. We conclude the section with a tems (MEMS) device [43]. Related concepts to that of discussion of emerging hybrid sensors in which correlat- flux concentrators are those of flux guides [44] (Fig. 3b) ing different sensor modalities may be particularly useful and flux transformers [45]. A flux guide enables the use when dealing with challenging environments. of a larger atomic OPM with high sensitivity for a smaller sensor area. This solves the issue of reduced sensitivity of small cells. A flux transformer can be used to decouple A. Flux engineering the ac fields to be measured from dc or slowly varying fields and gradients in the measurement region that can Flux engineering provides a suite of signal enhance- adversely affect the performance of the sensor. This is a ment tools specific to magnetic sensing and includes particularly useful “trick” in the case of magnetic reso- flux concentrators, flux guides and flux transformers. nance imaging (MRI), where fields and gradients are used A flux concentrator utilizes a high-permeability mate- to polarize the nuclear spins and encode information on rial to compress the magnetic flux from a larger area their spatial location [45]. 5

FIG. 3. Flux concentrators used in conjunction with optically pumped magnetometers: in (a) to enhance the sensitivity to spatially homogeneous magnetic fields in NV-center magnetometry [39] and in (b) to increase the spatial resolution of a vapor-cell magnetometer [44].

B. Magnetic shielding C. Signal modulation and filtering

Filtering is a simple solution to remove time-varying Ideally, to eliminate noise, one could shield all envi- noise in a frequency band that has minimal overlap with ronmental magnetic fields from the sensor. Metals with the signal band. Filtering can be utilized to remove both high relative magnetic permeability (mu-metal) can be magnetic noise or non-magnetic noise which presents as shaped and layered to form shielding which can effec- a magnetic signal. As discussed above, examples of envi- tively block out magnetic fields at all frequencies, cre- ronmental transient magnetic noise sources include power ating an ideal environment for magnetometers to focus lines, trains, elevators, communication signals, as well as on a signal of interest. For human-size experiments, the noise of the geomagnetic field. Noise sources are also typ- state of the art is the BMSR-2, an eight-layer magneti- ically present in the sensor apparatus; for example, the cally shielded room at the Physikalisch-Technische Bun- stability of the applied bias field, the optical excitation desanstalt (PTB) in Berlin with a passive shielding factor intensity, and RF control amplitude and phase. Stan- of 106 at frequencies >0.01 Hz and with a typical residual dard techniques to filter this noise is to include notch magnetic field < 1 nT [46]. One critical component for filtering, adaptive notch filtering [52, 53], independent performance is degaussing the shielding layers. With an component analysis [54], variation of the interrogation improved degaussing system with distributed coils, resid- time relative to the noise period, reference measurements ual fields as low as < 100 pT are theoretically possible on timescales shorter than the noise correlation time (e.g. with only a three-layer room and residual fields <130 pT used in Ref. [55]), and “phase cycling” as most frequently over a (0.5 m)3 cubic enclosure have been realized [47]. discussed in the nuclear magnetic resonance (NMR) con- Nanotesla-level shielding is typically utilized in funda- text [56]. There are, however, clear limitations to filter- mental physics experiments (e.g. the search for electric ing. Knowing the nature of the noise does not always dipole moments of atoms, molecules, and the neutron) allow one to simply “filter it out”. In measuring bio- and for biomagnetic measurements such as neural imag- magnetic transients, for example, a 50 Hz (60 Hz in the ing utilizing magnetoencephalography [47–49]. USA) notch filter over an action potential lasting 10 ms As magnetometers themselves become smaller and would cause significant distortion of the physiological sig- more portable, so should the shielding solutions. Tran- nal, since any band limitation in the frequency domain scranial magnetic stimulation (TMS)-evoked muscle ac- creates artifacts such as ringing, phase lag, or signal at- tivity has recently been detected inside a hospital set- tenuation in the time domain [57, 58]. ting [50] utilizing a portable shield which only encom- Filtering is often combined with signal modulation, passes the forearm and hand of the subject [51] (Fig. 4). which shifts the signal from a noisy low-frequency band This application highlights the utility of small, compact to a higher-frequency band to enable the use of notch sensors and shields. TMS requires a high-intensity mag- filtering or lock-in detection method. Signal modulation netic pulse applied to the brain which must be shielded techniques relevant for OPM devices are modulations of for the detection of 10 pT level signals at the hand. the optical excitation [21] (amplitude, frequency and po- 6

higher frequency. In one novel upconversion method, the NV-center in diamond is used to detect a static trans- verse magnetic field via the detection of the oscillating field due to the Larmor precession of a nearby nuclear spin [62]. One potential advantage of detecting upconverted ac signal is that one can utilize rephasing techniques to ex- tend the spin coherence and thus improve the sensitivity of the sensor. For ac magnetic sensing, the sensitivity ∗ ∗ is limited by T2 > T2 . Here, 1/T2 and 1/T2 are the homogeneous and inhomogeneous relaxation rates of the spins, respectively. The ac magnetic sensitivity is further enhanced with dynamical decoupling pulse sequences, pi- oneered by the NMR community, used to rephase the magnetometer spins [63]. Alternatively, this rephasing technique can be described in the language of filtering [64] or quantum lock-in amplification [65]. Enhancements up ∗ T2/T2 could be naively expected, however often the same ∗ noise fields that limit T2 will also be upconverted [62]. The advantage of enhanced interrogation time however remains. While spin-rephasing techniques are commonly used in solid state magnetometry, recently, the use has been extended to alkali atoms in solid parahydrogen [66].

D. Common-mode-noise rejection

Common-mode noise can be filtered utilizing more than one measurement. This strategy exploits a depen- dence on magnetic field between the two measurements which differs either in sign or magnitude from other noise sources. Here we discuss some specific examples of this strategy: differential magnetometry, gradiometry, net- works of sensors, double-quantum and comagnetometry.

1. Differential magnetometry, gradiometry, sensor networks

Differential measurement is perhaps the most common FIG. 4. (a) Biomagnetic measurements in a hospital environ- approach to dealing with noisy environments [69]. Sup- ment with partial magnetic shielding. The recorded magnetic pose, for instance, one needs to make a measurement of signals in the subject’s hand inserted in a shield are triggered a local source (e.g. human heart or brain). The mag- by transcranial magnetic stimulation. (b) Several commercial netic fields produced by local sources fall rapidly with atomic zero-field OPMs (in blue casing) are used to measure distance r from the source (as r−3 for a source that can the spatial distribution of biomagnetic signals from a stimu- −2 lated muscle. be approximated as a point dipole or r for a linear dipole), while unwanted fields from distant sources pos- sess a much larger degree of “spatial coherence.” If we have a sensor next to the source of interest, ideally, at a larization), RF frequency [39], magnetic field [59], and distance no greater that the spatial extent of the source, flux amplification [42] (Sec. III A). Mechanical rotation then the signal on the proximal sensor is a sum of the of the entire diamond sensor is another potential mod- signal of interest produced by the source and a signal ulation technique [60] (Fig. 5). In some cases, such as from spurious distant sources. Because of the larger spa- materials characterization [61], the sample itself can be tial extent of the latter, that part of the signal can be rotated or vibrated in order to move the signal into fre- subtracted using the signal from a second, reference sen- quencies where 1/f noise is reduced. Indeed, frequency sor at a distance much larger than the spatial extent of modulation is often referred to as frequency “upconver- the source. In the case where the field of interest and sion” since the low-frequency signal is converted to a the background both vary in time sufficiently slowly, the 7

tance between the cells. The use of the same probe beam to interrogate the two cells has a benefit of reject- ing common-mode noise related to the probe laser beam (this could be useful also for non-gradiometer measure- ments, as recently demonstrated, for example, with a so- called light-shift dispersed Mz sensor [71]). The reported gradiometric√ sensitivity of this device was better than 18 fT/cm/ Hz, sufficient for reliably detecting biomag- netic signals from a human body in unshielded environ- ment. In fact, OPM gradiometers have already been used to detect human brain signals outdoors [72], as shown in Fig. 6. A single-cell intrinsic√ magnetic gradiometer with a sensitivity of 10 fT/cm/ Hz was reported in [73].

FIG. 5. Diamond-sensor rotation rates exceeding 3000 kHz have enabled T2-limited sensing of a dc field [67] and rotation sensing [68]. (a) Diamond on a drill (Image credit A. Wood / U. Melbourne) (b) Measuring rotation with a diamond. From Ref. [60] . second sensor may not even be needed if it is possible to move a single sensor to alternate the source plus back- ground and background measurements. A logical continuation of this approach is a magnetic FIG. 6. Field test of an OPM designed for recording mag- gradiometer–a device that is insensitive to a uniform netic signals from the brain [72]. Professor M. V. Romalis field, but rather measures the spatial variation of the of Princeton University was a volunteer subject for the test. field. A “synthetic” gradiometer can be constructed from Photo courtesy Dr. Tom Kornack (Twinleaf, twinleaf.com). separate individual sensors, and, in some cases, it is pos- sible to construct a sensor that is intrinsically a gradiome- With multiple sensors, one can set up a high-order gra- ter. An example of a modern intrinsic gradiometer used diometer, and, as a further extension of the idea, an ar- for brain measurements in an unshielded environment ray of sensors (for instance, as a wearable brain senor, (see Sec. V) is that described in Ref. [70]. The authors see Ref. [74] and references therein) and even a magne- constructed a scalar gradiometer using two miniature tometer network. Such networks are capable of detecting (5×5×5 mm3 inner volume) cesium vapor cells separated signals with complicated spatio-temporal signatures and by 5 cm, and vertical-cavity surface-emitting lasers (VC- find applications in geophysics [75], fundamental physics SELs) producing independent pump and probe beams [76, 77], and even in the study of magnetic signatures of for both cells. Off-resonant linearly polarized probe light cities (Sec. IV K). It should be pointed out that extraction interrogated the two cells at the same time, so that the of maximal amount of useful information from a network output of the gradiometer was proportional to the change of sensors is generally a highly nontrivial task (see, for ex- of the magnetic field over the baseline given by the dis- ample, Ref. [77] discussing the problem in the context of 8 the Global Network of Optical Magnetometers for Exotic (a) Physics Searches (GNOME); see also Sec. IV K) and ben- efits from modern data-analysis techniques like machine learning.

2. Double-quantum magnetometry

Double-quantum (DQ) magnetometry utilizes a pair (b) of spin transitions whose frequency response is identi- cal to most non-magnetic perturbations but differs for magnetic signals. For the specific case of NV centers in diamond, these transitions correspond to the “single- (c) (d) quantum” ms = 0 ↔ ms = −1 and ms = 0 ↔ ms = +1 transitions (Fig. 1(g)). The two transitions have iden- tical dependence on electric field, strain, pressure, and temperature, while an equal but opposite dependence on magnetic field (and rotations about the NV axis) [78–81]. Thus by subtracting the two single-quantum signals, fluc- tuations in the non-magnetic field are eliminated while magnetic sensitivity is enhanced (see, for example, [82] in which this technique is used wide-field magnetic imag- ing of hemozoin crystals). In a similar vein, one can add FIG. 7. (a) The use of dual-frequency pulses in double quan- the two signals to obtain enhancement of a non-magnetic tum magnetometry eliminates noise due to temperature and signal (for example, for thermometry [83–85]). electric field (reproduced from Ref. [87]). (b-d) The use of It is often advantageous to directly create a “double- four-frequency pulses (f1tof4), biased on the outer and inner edges of the single quantum resonances, enables double quan- quantum” (DQ) superposition of the ms = ±1 states. The original double-quantum demonstration utilized a tum magnetometry in imaging applications. (From Ref. [90]). In this imaging application, a CMOS camera is used with a single-tone, broadband RF pulse to create this superpo- 4 ms exposure time. The pulse sequence (b) thus runs many sition [86]. More recent demonstrations utilize two-tone times during a single exposure. (c) energy level diagram. (d) pulses [87] (Fig. 7a) or multiple pulses [88] that have the bias points in the NV ODMR spectrum. advantage of allowing operation at arbitrary bias fields. The DQ technique can be also implemented in a cw mode, in which the two-tone microwaves are modulated, and between magnetic field and rotation-induced frequency detection is done with a lock-in method [39]. DQ mea- shifts is thus of significant importance. A solution to surements have demonstrated a two-fold enhancement in this challenge is provided by comagnetometers [96]. Co- signal-to-noise [87] and a 15-fold improvement in mag- magnetometers compare the spin precession frequencies, netic sensitivity by eliminating thermal drifts [86] rel- γ1,2B + Ω, between two species with different, known ative to single-quantum magnetometry. The DQ tech- gyromagnetic ratios γ1,2. Ω is the rotation rate. Co- nique was also utilized to eliminate strain gradients and magnetometry variants include utilizing different atomic inhomogeneities that present as spurious magnetic sig- species [97] and different isotopes [96, 98, 99] and hyper- nals [89]. Finally, four-tone RF pulses, Fig. 7b) were fine levels of the same atomic species [100, 101]. The utilized in wide-field imaging applications, in which one latter implementation is particularly robust to system- frequency is used on each side of the two electron res- atic errors due to magnetic field gradients. When imple- onances (Figs. 7 c,d). This “double” DQ magnetometry menting a comagnetometer, a concern is co-locating the eliminates the effect of inhomogeneity of both the res- two species; otherwise, the measurement can be compro- onance position as well as the lineshape across the im- mised. One approach for nuclear spin-based sensors is to age [90]. use the spins of mixed liquids [102] or even different spins residing in one and the same molecule [103].

3. Comagnetometry E. Hybrid modalities The resonance frequency of a spin is shifted when the sensor undergoes rotation. While this effect is the basis Challenging measurements may benefit from hybrid for rotation sensors, it can also be a systematic error for modalities, that can both enhance the signal of inter- magnetometry, and is especially relevant for moving sen- est and discriminate noise. We divide hybrid modalities sor applications (Sec. IV H) and spin-based fundamental- into two categories. physics experiments [91–95]. The ability to distinguish In the first category are sensors that detect multiple 9 complementary physical signatures of the investigated IV. PHYSICALLY CHALLENGING object or phenomenon. Examples include the combina- ENVIRONMENTS tion of wide-field optical microscopy and NV magnetom- etry [104, 105] used to realize NV-based force-induced In this section we focus on the physical challenges pre- remnant magnetization spectroscopy (NV-FIRMS) for sented in practical, high-sensitivity magnetometry. One measuring the magnitude of binding forces between bio- of the major applications of magnetometry is the study logically relevant molecules, as well as wide-field NV mag- of materials [111, 112]. For these studies, sensors must netometry combined in a single setup with a magneto- be able to operate in extreme temperatures and pres- optical Kerr-effect (MOKE) microscope [106]. The idea sures. Size also presents a physical challenge-whether of the latter is that MOKE is sensitive to surface magne- probing a nanoscale phenomenon or astronomical ob- tization which does not necessarily produce a magnetic jects. Practical deployment often requires operation on field outside of the sample, so combining it with NV mag- moving platforms requiring sensors robust to vibrations netometry that measures the field provides a more com- and changes in orientation with respect to the Earth’s plete diagnostic picture. (relatively) large magnetic field. Finally, space and ac- In the same category is also real-time tracking of a pa- celerator applications require sensors robust to radiation. tient’s movement using video cameras, enabling magne- While the focus of this section is the robustness of the toencephalographic recording without sedating the sub- sensor’s physical magnetic transduction mechanism, we ject [107]. Eliminating the need for sedation is partic- note that the sensor must include control and readout ularly important when working with children. Comag- components. Some of the novel methods to eliminate lo- netometers using different species (Sec. III D 3) are, in a cal controls and further enhance robustness to extreme sense, also in the same category of hybrid sensors. physical environments include all-optical and microwave- free magnetometry which are discussed in Sec. II A, II B. In hybrid sensors of the second category, the physical parameter being measured (e.g., position, pressure, tem- perature) is “transduced” into another parameter (e.g., A. Temperature challenges magnetic field, frequency of optical resonance) and mea- sured via sensing the latter. A striking example of such a hybrid sensor is a nanothermometer composed of NV cen- Temperature can dramatically change material prop- ters and a magnetic nanoparticle (MNP) [108]. This de- erties and magnetic phase changes can be observed from vice takes advantage of the ferromagnetic-paramagnetic cryogenic temperatures to 1000 K. Additionally, biosen- transition of the nanoparticle material that occurs at sitive measurements can only be performed in a limited a certain temperature. The temperature susceptibility temperature range. If a sensor itself has a limited oper- of the NV magnetic resonance in this hybrid device is ational temperature range, temperature isolation of the 14 MHz/K, an enhancement of over two orders of mag- sensor can be utilized to extend the complete sensor pack- nitude over the temperature susceptibility of NV centers age operational range. Examples of this include the use of SQUIDs (requiring cryogenics) in magnetoencephalog- themselves [109]. The demonstrated√ sensitivity of the hybrid nanothermometer was 11 mK/ Hz under ambient raphy [49] and hot vapor cells for measurement of biolog- conditions. The authors of Ref. [108] envision that hybrid ical [113], geological [3, 61] and chemical [114] samples. thermometers can be designed to operate from cryogenic Thermal isolation is also an important consideration for temperatures to about 600 K by tuning the chemical com- keeping the overall power consumption of the device low. position of the MNP. However, this isolation comes at the unavoidable cost of reduced spatial resolution and experimental complex- In another type of NV-magnetic nanoparticle hybrid ity. Thus sensors which function over a wide tempera- sensor, the nanodiamond NV magnetometer and MNP ture range are particularly attractive. Optically-pumped are connected by a stimulus-responsive hydrogel spacer magnetometers by definition work far from equilibrium, (Fig. 9f) that changes its length as a function of temper- and for solid-state OPMs this feature enables them to ature, pressure or power of hydrogen (pH) of the environ- function from cryogenic temperatures all the way up to ment, leading to a change of the magnetic field detected 700 K [8, 62, 115]. Vapor-cell magnetometers are indi- by the NVs [13]. rectly sensitive to temperature as they require a certain Similar ideas can also be used in fundamental-physics and stable alkali vapor pressure (and therefore number experiments. A recent example is Ref. [110], where the density) to perform. Atomic magnetometer in the spin- authors search for exotic interactions between two elec- exchange relaxation free regime even require stable tem- tron spins, that are separated by “molecular rulers” of peratures above 400 K to perform [116]. certain lengths, and interrogated via a double electron- One clear advantage of NV-based magnetometers is the electron resonance (DEER) technique. Comparing the diamond host. Diamond is stable in air up to 1000 K [117] DEER signals for different-length separation between the and can withstand higher temperature in an inert gas spins allows one to search for any deviations from the atmosphere [118]. The NV center is stable at tempera- usual dipole-dipole interaction between the spins on the tures exceeding 1350 K [119]. However for all optically- nanometer scale. pumped magnetometers, the spin and optical proper- 10 ties change with temperature, limiting the operational was further refined and applied to imaging of magnetiza- range. At room temperature, the NV longitudinal spin tion in monolayer (i.e., two-dimensional) van der Waals relaxation time T1 exceeds 6 ms and increases at lower magnets [137]. Current work aims to extend the tem- temperature [120]. While the spin-coherence time and perature range of scanning NV based magnetometry into thus magnetic sensitivity can approach the T1-limit uti- the millikelvin range [138]. lizing dynamical decoupling techniques [121], dynami- cal decoupling is only suitable for small-bandwidth ac sensing applications (Sec. III C). Thus typically at cryo- C. High pressure genic and room-temperature, the NV dc sensitivity is limited by noise due to paramagnetic impurities. Room- Like temperature, pressure can also dramatically temperature operation extends NV magnetometry appli- change material properties [112]. Magnetometry can elu- cations to biological systems which are discussed further cidate important aspects of these changes in studies of in Sec. V. phase transitions and material synthesis [139, 140], and in The longitudinal relaxation rate of NV sensors is a the exploration of new superconducting materials. Geo- nontrivial function of the temperature [122], including physicists can simulate and study magnetic properties of hitherto unexplained “plateau” of nearly constant relax- the Earth’s core [141], or understand the formation of ation rate between a few K and up to 100 K or higher different rock types [142], while chemists can study pres- depending on the density of the color centers in the sam- sure induced changes in the electronic structure of atoms ple [120, 123]. At temperatures T>500 K, T1 for NV and molecules. centers is found to be proportional to T−5.6 with T 1 State-of-the-art diamond anvil cells (DACs) can rou- ranging from T 516 K = 119 s to T 944 K = 3.8 s [124]. 1 µ 1 µ tinely achieve 100s of GPa, and pressures exceeding In practice, at high temperature magnetic sensitivity 1 TPa have been reported [143]. Most magnetometers is again not limited by T . Instead, high-temperature 1 cannot withstand high pressures and thus sensing is per- sensitivity is limited by non-radiative processes limiting formed outside of the high pressure environment. Two optical-spin readout contrast and also typically restrict- examples of these ex situ sensors are SQUID-based and ing NV magnetometer operation to temperatures below NMR-based magnetometry. In SQUID-based measure- 700 K [8, 115, 125]. The spin coherence, however, is still ments, a superconducting coil is placed around or near robust at higher temperatures and single-NV optically the vicinity of the sample [140, 144]. While this approach detected magnetic resonance (ODMR) using a pulsed utilizes the excellent magnetic sensitivity of SQUIDs, the heating technique was utilized to measure the phase tran- cryogenic requirements place constraints on both the pos- sition in a single Ni nanoparticle at 615 K [124]. sible temperature range of the samples, and the materials of the high-pressure cell [145]. Furthermore, the diame- ter of the induction loop is much larger than the sample B. Cryogenic environment size, and thus does not provide magnetic flux resolution. NMR techniques, which directly probe spin-spin inter- Conventional vapor-cell OPMs cannot operate in the actions in materials, use an induction coil and gradient cryogenic environment, though there have been efforts bias field [146–148]. Induction-coil NMR techniques miti- towards magnetometry with paramagnetic atoms in cryo- gate the cryogenic constraints and, while not yet utilized genic solids, e.g. [126], liquids, e.g. [127], and gases, e.g. in DAC experiments, can achieve sub-mm spatial reso- [128, 129]. lution. However challenges include poor filling factors, As discussed above, diamond sensors can operate at sample alignment and the presence of large background low temperatures, and even benefit from prolonged re- fields. laxation times [121] at cryogenic temperatures. One low- NV spin resonance were observed up to 60 GPa [9] temperature application has been the study of condensed and, in contrast to the above techniques, an NV mag- matter phases [130–134]. Another possible application is netometer in diamond can be placed in close proximity monitoring fields of superconducting magnets operating to the sample volume (Fig. 8a), even directly integrated at 4 K. Compact and robust absolute sensors for this ap- into the diamond anvil [149–151], and spatial resolution plication can perhaps be constructed based on the effect can be optically diffraction-limited. A comparison of spa- of electromagnetically induced transparency in, for exam- tial resolution and magnetic sensitivity for high-pressure ple, SiV centers in diamond, enabling all-optical sensing, magnetometry is given in Fig. 8b. Recently integrated see Sec. II A. NV-DAC systems were utilized to investigate pressure- Scanning magnetometry (see e.g. Fig. 9a) is im- induced phase transitions in iron [149, 150] (Fig. 8c) and portant for materials research and condensed-matter Nd2Fe14B [152], the pressure-temperature phase diagram physics [111, 135]. A 4 K scanning NV magnetometry of gadolinium [150], and the superconducting transition experiment [136], which allowed quantitative character- temperature, critical fields and the Meissner-state local ization of the nanoscale structure of a vortex in a su- magnetic field profile of BaFe2(As0.59P0.41)2 [151]. These perconductor, provided a proof of concept for scanning investigations have focused on high-pressure dc magne- magnetometry in cryogenic environment. This technique tometry; however the extension of NV-based NMR de- 11 tection [153–156] to DAC systems should enable spatially In the highest-field pulsed magnets [163], the field resolved NMR studies [157]. is often measured via magneto-optical (Faraday) rota- tion in a sample of a solid transparent material such as glass. Faraday-rotation measurements are largely D. Low pressure immune to electromagnetic pick-up accompanying the magnetic-field pulse and allows measurement of the field Similar to other solid-state magnetometers, and in con- evolution on fast (µs) scales with large dynamic range. trast to atomic vapor magnetometers, low-pressure op- eration of NV-based magnetometers is straightforward. The aspects of low-pressure magnetometry unique to the F. Near-zero magnetic fields NV (or similar defect-based systems) include nanoscale scanning probe imaging [158] and the potential abil- Certain applications including searches for exotic inter- ity to combine low pressure with a large temperature actions (e.g., those due to electric-dipole or gravitational- range [111]. dipole moments of particles, see [164]), biomagnetic mea- surements in shielded rooms (Sec. V), or detection of zero- to ultralow-field (ZULF) NMR [165] require op- E. High magnetic fields erating magnetic sensors, including gradiometers [166], at near-zero field. Indeed, some sensor types, for exam- The techniques to measure high magnetic fields (tens ple, the most sensitive alkali vapor based spin-exchange of tesla and higher) were recently reviewed in Ref. [159], relaxation-free (SERF) OPM (see, for example, [116]) so we only provide a brief summary here. typically require operation at fields below 10−7 T. On the The strongest DC fields obtained with laboratory mag- other hand, it was originally thought that diamond NV nets are currently around 45 T [160]. With pulsed mag- center based sensors lose first-order sensitivity to mag- nets that are not destroyed during a pulse, one can go netic fields at fields below ≈ 0.1 mT due to the mixing to slightly over 100 T. “Explosive” techniques reach to of Zeeman sublevels by electric fields and strain inter- over 2 kT. Clearly, different techniques are required for nal to the diamond crystal. To overcome this limitation, measuring magnetic fields in each of these settings. one approach realized for a near-zero-field NV-ensemble For DC magnets, NMR teslameters are typically the sensor [167] is to use, instead of an external bias field, instruments of choice with the highest relative accuracy the hyperfine magnetic field provided by the nitrogen of better then a part in 1012 achieved with magnetome- nucleus of the NV center. For this to work, one needs ters based on gaseous 3He [161]. The absolute accuracy a low-strain, narrow-line (e.g., 13C depleted) sample and is limited to approximately a part in 108 by the finite polarization-selective microwave excitation. In Ref. [167], accuracy of fundamental constants. The operation range circularly polarized microwaves were used to address spe- of 3He magnetometers is limited to about 12 T due to the cific hyperfine transitions in ODMR. The authors also effect of “magnetic decoupling” of the hyperfine structure describe how applying a modulated magnetic field allows (leading to the nuclear spin quantum numbers being con- one to perform zero-field magnetometry with linear po- served in optical transitions) that reduces the efficiency larized microwaves. Microwave-free zero-field ensemble of optical pumping. magnetometry [168] and zero-field magnetometry with a Magnetic fields up to 45 T are measured with Faraday- single NV center [169] have also been demonstrated. induction pick-up coils (during the ramp-up of the field in the magnet) that are calibrated with NMR “marker” lines (see [159]). NMR marking and Faraday-induction G. Challenging spatial scales pick-up coils are also used for pulsed magnetic fields with typical field-evolution times in the ms to tens of ms range. Magnetic measurements are performed at spatial scales Calibration can be accomplished using the de Haas - van from atomic to astronomical, see Fig. 9. In this section, Alphen effect in copper wherein the magnetization of the we discuss some of the extremes. material oscillates as a function of magnetic field due to the Landau quantization of electron energy. The highest-precision metrology of pulsed magnetic fields is based on the use of rubidium in gas phase [162]. 1. Milli, micro, and nanoscales The magnetic field “tunes” the energies of the Zeeman sublevels in the upper and the lower state of an atomic The size of a magnetometer limits both its spatial res- transition and brings the transition on resonance with olution and its integration into physical systems. When light of a fixed and precisely measured frequency, in- it comes to compact sensors, optically-pumped magne- creasing absorption and fluorescence. The instrument tometers stand out. First, the sensing element can be of Ref. [162], based on the 780 nm (D2 line) in rubidium, small or even nanoscopic; NV emission has been ob- demonstrated an accuracy of ≈ 2 · 10−4 for the entire served in single-digit nanometer diameter nanoparticles duration of the pulse. produced by milling [170], detonation [171, 172], and ox- 12

(a) (b)

(c)

FIG. 8. (a) Schematic of NV magnetometry measurement in a high-pressure diamond anvil cell from Ref. [9]. (b) Comparison of high-pressure magnetometry techniques from Ref. [150] (c) Maps of the ODMR frequency splitting in a diamond sensor above an MgB2 sample at 7 GPa, reproduced from Ref. [149]. Below 30 K, the exclusion of the magnetic field from the sample is observed. idative etching [173]. Second, optical and RF control ily limited by the defect-surface distance. A new planar eliminates the need for local electrical control. scanning microscopy modality was recently introduced Nanoscale: ODMR is observed in diamond particles of [189], in which there is no tip at all, and a macroscopic several nanometers in size [170], however practical mag- diamond plate containing a shallow NV is scanned par- netometry with such diamond remains challenging due allel to its surface. NV scanning probe sensors were to the high NV density required to ensure there is a used for high-resolution imaging of a multitude of phe- color center in such a small nanoparticle [174]. Another nomena ranging from antiferromagnetic domains [186] to issue is degradation in both photostability [174, 175] skyrmion bubbles [190] and coherence due to paramagnetic surface states [176, Beyond scanning probe applications, nanostructured 177]. Promisingly, advances towards addressing these diamond has additional features including an increased challenges continue, including increased NV incorpora- surface-to-volume ratio that has been leveraged to en- tion [178], advanced surface functionalization [179], and hance solution NMR sensitivity [191] and light-guiding top-down fabrication of larger nanoparticles from bulk di- capabilities for enhanced photon collection [192, 193]. amond [180]. Nanoparticle NV-magnetometers are par- Similar to nanoparticle diamond, in nanostructured dia- ticularly suitable when the nanoparticle sensor is func- mond, charge stability, photostability, and spin coherence tionalized to the sample (Fig. 9f). For example, com- times are still outstanding challenges, however the single- posite diamond and magnetic nanoparticle sensors can crystal surface and control over the NV placement have be utilized for enhanced temperature sensing in hybrid enabled further studies towards addressing these prob- sensors discussed in Sec. III E. lems [176, 194, 195]. Nanoparticles were also utilized in the first nanoscale Close proximity between sample and sensor can also NV scanning probe experiments [181, 182], however most be achieved utilizing ensembles of NV centers synthe- scanning probes today utilize nanoscale diamond tips sized in a planar sheet at the diamond sample surface. fabricated from bulk diamond (Fig. 9a). High spatial res- This is for example used in Ref. [201] to image oscillating olution is enabled by the creation of NV centers nanome- magnetic fields in the microwave regime with high spa- ters from the single-crystal-diamond surface [183, 184] tial resolution (Fig. 9i). Vapor-cell magnetometers are followed by reactive ion etching of the diamond [11] to normally much larger but can also be produced to ex- define the tip. Several commercial companies now sup- hibit at least one dimension down to 40-1000 nm as in ply these tips for scanning probe systems (e.g. Qnami the case of nanocells [203]. The reduced sensitivity to AG, QZabre-LLC). State-of-the-art spatial resolution is spatial scales is offset by the better signal-to-noise ratio typically in the 25-50 nm range [185–188] and is primar- of vapor-cell magnetometers, so they can still be used 13

FIG. 9. Magnetometry at different length scales spanning 16 orders of magnitude. (a) diamond scanning-probe tip, anti- ferromagnetic domains imaged with a scanning single-NV probe [196]; (b) chip-scale atomic magnetometer from [197]; (c) a typical-size vapor cell [1]; (d) telescope launching a laser guide star for magnetometry with sodium in the mesosphere [198]; (e) satellite-based magnetometry [199]; (f) nanodiamond hybrid sensor from [13]; (g) magnetic images of superconductor vortices obtained with NV centers [133]; (h) 100 µm NV ensemble sensor from [200]. (i) Wide-field imaging of oscillating magnetic fields in the GHz range with NV centers [201]; (j) unmanned aerial vehicle fitted with vapor OPMs for magnetic surveying from [202]; (k) magnetic anomalies detected with the devices in (j). to detect, for example, sub-picomolar concentration of Compact weight and size also is advantageous for satellite magnetic nanoparticles [204]. In diamond, planar sheets deployment (Sec. IV G 2). of NV centers have enable wide-field imaging magnetic phenomena such as eddy currents [205], magnetotactic bacteria [206], and fluxtubes on a superconductor with 2. Space-borne and remote magnetometry a wide-field setup [133] (Fig. 9g). If the close environ- ment of the sensor is (super)-conductive, eddy currents in At the opposite end of the size continuum is space- the material attenuating the delivery of oscillating mag- based magnetic sensing [213]. Magnetometry can pro- netic fields reduce the sensor’s performance. Mitigation vide powerful information on the interactions between strategies include all-optical and microwave-free sensing solar wind and solar system bodies and can further our protocols discussed in sections II A, II B. The sensitiv- understanding of planetary interiors. There are two main ity to the electromagnetic properties in the sensor’s local approaches: bring earth-made sensors into space (see e.g. environment can and has been turned into an advantage Fig. 9e) or utilize atomic vapors already present near a to perform material-characterization studies with both heavenly body of interest. vapor cell sensors [207–211] and diamond based sensors [205]. The competitive edge of optically pumped mag- Atomic-vapor magnetometers have been employed in netometers with respect to inductive eddy-current detec- spacecraft missions since the 1961 Explorer X satellite. tion for nondestructive evaluation is the high sensitivity An overview of NASA’s missions employing atomic vapor at low frequencies (high conductivity) and the high spa- magnetometers is found in Ref. [213]. The key consid- tial resolution given by the potentially small sensor size erations for satellite deployment include weight, volume and near-field measurements. (Sec. IV G 1), cost, sensitivity and resilience to tempera- ture changes (Sec. IV A) and radiation (Sec. IV I, as well Micro to milliscale: The above materials challenges as power consumption. Atomic magnetometers readily in nanoscale NV magnetometry are less of a problem satisfy the dynamic range requirements from nT to tens for micron-scale and larger devices. And for millimeter- of µT for interplanetary and planetary exploration. The scale applications, chip-scale atomic vapor magnetome- principal advantage of atomic magnetometers over the ters are now possible due to recent advancements in sili- more compact and low-cost fluxgate vector magnetome- con chip-based atomic vapor cells [197] (Fig. 9b). The in- ters is the absolute nature of atomic sensors which do not tegration of micro-to-mm sensors and sensor arrays [212] require calibration. This property has led to a proposal into biomedical devices is expected to impact an array to utilize atomic magnetometers in the evolved Laser of biomedical applications as discussed further in Sec. V. Interferometer Space Antenna (eLISA) designed to test 14 low-frequency gravitational wave radiation [214]. eLISA channels or a porous metal [224]. Here, water is pre- requires low-frequency, high-sensitivity magnetic sensors polarized in a strong magnetic field, then flows into the because of the magnetic nature of the test masses in object to be imaged, where the spatial information is en- eLISA; a significant source of expected acceleration noise coded into the time dependence of the polarization of the will be magnetic noise. flowing sample, which is finally detected with a magne- Due to the compact size, broad temperature operation tometer downstream. Remote-detection NMR be useful range, radiation resistance and ability for photo-electric for imaging large objects or when one cannot get the sen- readout, diamond-based OPMs are also taking their first sor into the object. The trade-off is that one cannot use steps into space [215]. They may eventually compete the more common and time-efficient frequency-encoded with various solid-state flux-gate [216] and magneto- NMR in this case and is reduced to using phase encoded, resistance (impedance) [217] sensors whose drawback is which generally leads to much longer image-acquisition that they are generally non-absolute and require calibra- times. tion. Remarkably, it is even possible to detect magnetic fields in near space via the gas already present. For example, meteors deliver elements to the upper atmo- sphere, and there is a layer of atomic sodium of sev- eral km in thickness that exists at a height of 90-100 km above the Earth surface. These atoms can be ex- cited with a beam of resonant light from a laser on the surface, creating a secondary (fluorescent) light source called a laser guide star (LGS). LGS systems using high- power lasers were developed for adaptive optics em- ployed for real-time compensation of atmospheric per- turbations and enabling near-diffraction-limited angu- lar resolution for surface based telescopes with diame- ters up to tens of meters. For LGS based magnetom- etry, the laser light is amplitude, polarization, or fre- H. Moving Platforms quency modulated in resonance with the Larmor fre- quency of the sodium atoms in the mesosphere, and the resonance is detected via a change in the intensity of laser-induced fluorescence [218]. Magnetometry with mesospheric sodium is an example of “remote magne- As OPMs become more compact, there emerge appli- tometry”, to highlight the remote detection scheme of cations on moving platforms, including handheld units, the highly local sensor. Several years after the original drilling assemblies [225], unmanned aerial and underwa- 2011 proposal [218], LGS magnetometry has been exper- ter vehicles (UAV, UUV), satellites (Sec. IV G 2), cars and trucks, submarines and aircraft (Fig. 10). Moving imentally demonstrated√ [198, 219, 220] with a current sensitivity of 30 nT/ Hz [198] (Fig. 9d). Long-term ap- platforms pose challenges due to “platform fields,” vibra- plications include mapping the Earth’s lithospheric mag- tions, accelerations, and “heading errors” (see below). netic fields, monitoring magnetic disturbances, long-term measurements on ionospheric currents and satellite-based explorations of other planets/moons with sodium atmo- When a magnetometer is operated in a passive mode, spheres. its optimal operation requires careful adjustment of the There is another version of remote or “stand-off” mag- parameters, especially the “working point” of the mag- netometry, where most of the magnetometer apparatus is netic resonance. There are several strategies that can be at at a location spatially removed from the location where combined to minimize the effect of vibrations on optimal the field is measured (as in LGS magnetometry), except operation including vibration isolation, monolithic inte- for just the sensor element (e.g., an alkali-vapor cell) and gration [197, 200, 226] (Fig. 9h), filtering (Sec. III C), a retro-reflector for the laser laser light (see, for example, closed-loop frequency-locking techniques [227], and de- Ref. [221]). Such schemes, including LGS magnetometry, vice designs that minimize/eliminate permanent or in- are examples of all-optical magnetometers discussed in duced magnetic fields from the sensor body. In the worst- Sec. II A. case scenario, the sensor may completely lose the reso- An interesting sensing application is that of remote- nant signal. Loss of resonance is of particular concern for detection NMR [114, 222, 223]. Here, the three NMR high-sensitivity, low bandwidth sensors. In such a situa- steps, polarization of the nuclei, encoding of the spatial tion, a self-oscillating device, see, for example, Ref. [228] or spin-composition information, and signal readout can and references therein, can be of practical advantage as it all be separated in space. This could be useful, for in- automatically restarts the optimum operation following stance, for imaging water flow inside a network of metal a strong perturbation. 15

Needless to say, OPMs do not hold a monopoly on operating on moving platforms. For example, an under- water SQUID based tensor gradiometer was recently de- ployed for scanning sea floor [237].

I. Radiation

Sometimes, magnetic measurements need to be per- formed in environments with high levels of radiation, for example, in space, in the vicinity of or within nuclear reactors [238], or at the sites of nuclear accidents such as the one that occurred in 1986 in Chernobyl, Ukraine. Other examples of high-radiation environments are the beam tunnels and interaction halls at particle accelera- tors (see Sec. IV J), as well as fusion reactors [239]. Because materials can be strongly affected by radia- tion, operation in high-radiation environments requires a particularly careful choice of materials and compo- nents used in a magnetometer and, ideally, extensive pre- operation testing with the type, energy, and radiation dose corresponding to a particular application. For ex- ample, vapor cells, coatings, and optical fibers should not undergo radiation-induced darkening (see Fig. 11), elec- tronic components should maintain their integrity. For- FIG. 10. Flight test, in 2012, of a towed magnetic gradiometer tunately, thanks to the considerable body of materials re- prototype developed by Southwest Sciences, Inc. in collabo- search in the context of space exploration, nuclear energy, ration with UC Berkeley. Insets: a) gradiometer optical head and accelerator instrumentation, there exist databases of assembly; b) aerodynamic “bird” with the gradiometer head radiation resistant materials and components (for exam- mounted in the tail. ple, [242]). Diamond based magnetometers are of particular inter- Measurement errors which arise due to the relative ori- est for sensing in high-radiation environments because entation of a magnetometer and the magnetic field, typ- diamond is among the best radiation-resistant materials. ically the Earth’s magnetic field, are termed “heading It is for this reason that diamond detectors are widely errors” [2, 229]. In OPMs, heading errors arise because used in high-energy physics (see, for example, [243] and of nonlinear Zeeman splitting (due to mixing of differ- references therein). ent hyperfine states by the magnetic field) and the light shift (due to mixing of the excited and ground states by the optical field). One practical but restrictive solu- J. Magnetometry at accelerator facilities tion is to fix the sensor orientation with respect to the Earth’s magnetic field [230, 231]. Modern solutions for Magnetic metrology is essential for the design and op- cancelling the nonlinear Zeeman effect include utilizing eration of accelerators and particle detectors [244] and ac- double-modulated synchronous optical pumping [232], celerator based experiments often require precise and ac- using high-order polarization moments [233] free of non- curate magnetic-field measurements. An extreme exam- linearities, compensation with tensor light shifts [234], ple of this is the “g-2” experiment at Fermilab that aims spin-locking [235], and combining clockwise and counter- to precisely measure the anomalous magnetic moment clockwise circular polarization [229, 236]. of the muon. The magnetic-measurement system in this Magnetic-anomaly detection (MAD) is often con- experiment [245] employs 378 probes in fixed locations ducted with magnetometers on board aircraft or towed around a storage ring to constantly monitor magnetic- with helicopters (Fig. 10). However, in recent years field drifts. These NMR sensors need to be periodically these platforms have been increasingly superseded with calibrated, which is done with a trolley which goes around UAVs (Fig. 9j,k) and drones (see, for example, https: the 45 m long circumference and takes data every 2 to 3 //areai.com/altius-500-4/). In such systems, mag- days. The trolley-sensors themselves gets calibrated with netic sensors have to withstand accelerations up 100 g at precision absolute probes. The level of absolute calibra- launch, a challenge met by some of the modern vapor-cell tion using 3He NMR has reached a remarkable level of based OPM such as Twinleaf’s microSAM-2 total field 32 ppb [246]. magnetometer with no dead zones (https://twinleaf. With all the remarkable high-precision results coming com/scalar/microSAM/). from advanced accelerator laboratories, one may not im- 16

Apart from magnetic noise, other challenges include radiation (Sec. IV I) and poor climate control in large ex- perimental halls.

K. Magnetometry in urban environment

Cities with their numerous and diverse sources of mag- netic noise associated with power lines, surface and un- derground rail systems, automobile traffic, operating ma- chinery and buildings, are particularly challenging for magnetometry. On the flip side, performing magnetic measurements in the cities can provide unique informa- tion on urban patterns whose analysis could be useful for the understanding and optimization of energy utiliza- tion, environmental monitoring, or for threat detection via detecting of unusual magnetic activity due to struc- tural changes or clandestine activity. In the recent work of [249], a network of global positioning system (GPS) synchronized√ flux-gate mag- netometers with a 100 pT/ Hz sensitivity sampled at ≈4 kHz was used to study the magnetic environment in and around the city of Berkeley, CA, USA. It was found (see also the references within [249]) that the domi- nant magnetic-noise sources were associated with the Bay Area Rapid Transit (BART) system, and ways to iden- tify much weaker signals in the presence of this dominant signal based on wavelet analysis applied to the network data were discussed. This work is the first step towards identifying the “magnetic profile” of different cities.

V. BIOSENSITIVE MEASUREMENTS

A. Magnetometry at the cellular level

FIG. 11. The effect of radiation on magnetometer vapor cells Significant interest has arisen in the use of magnetom- [240]: (top) before irradiation, (middle) after 3.6 Mrad (a typ- etry, and specifically NV centers in diamond, to study ical dose for a mission to the Jovian moon Europa [241]), biological processes and biomolecular structure of living (bottom) after 360 Mrad. The cell in the middle has optics organisms. Many of the challenges for noninvasive sens- glued to it. ing in these systems arise from the scale of the sensor and apparatus, relative to the probed system, which is often at the cellular level. The relative simplicity of NV setups, mediately realize that magnetometry in such an envi- as compared to the large cryogenic apparatus required ronment may, indeed, be especially challenging. As is for SQUID systems, has made them attractive for use in well documented, for example, for the CERN Antipro- detection of biological signals, for example, detecting ac- ton Decelerator hall [247] and the neutron electric-dipole tion potentials in living systems, or probing bacterial cells moment (nEDM) experiment at the Paul Scherrer Insti- placed on a diamond surface with NVs implanted just be- tute [248], there are multiple sources of spurious mag- low. Such a setup allows for not only sensing, but optical netic fields with diverse signatures such as pulsed extrac- imaging of the magnetic field, providing detailed infor- tion magnets, overhead cranes, and various other activi- mation about production of magnetosomes in bacterial ties and devices associated with operation of the facility. cultures or in localizing magnetic markers or nanoparti- Here active and passive shielding, as well as monitoring cle tracers [206, 250]. Similar setups have been used to stray fields outside the experimental region are among perform magnetic resonance imaging of spin-labelled pro- the possible mitigation approaches. In the worst case teins in cell membranes [251]. In Ref. [252], it was shown scenario, the measurement needs to be temporarily in- that sensors could be placed as close as 10 µm from the terrupted while the perturbation is particularly intense surface of an undissected marine fanworm Myxicola in- [248]. fundibulum. At this distance, the magnetic signal as- 17 sociated with stimulated action potentials√ were as large cal perspective, magnetic fields offer an additional modal- as 1 nT, compared to the 15 pT/ Hz sensitivity of the ity by which to study the muscle and nervous system of sensor, demonstrating the unique combination of sample the body that is distinct from electric sensing techniques proximity and high sensitivity afforded by NV centers in and magnetic resonance imaging [263, 264]. diamond for biological applications. Whether originating from nerves or muscles, the de- The low cytotoxicity of diamond and robust fluores- tected magnetic signal is the sum of fields from many cence properties have motivated the use of nanodiamonds individual cells firing respective action potentials. The with NV centers as exquisitely sensitive, local probes of aggregate detected signal, nevertheless, remains within the fields even within these living systems, where the the reach of only the most sensitive of magnetometers challenges for magnetometry with SQUIDs, fluxgate sen- [262, 265]. The highest-amplitude magnetic signals from sors, or atomic vapor cells could be insurmountable [253– the body emanate from the heart, with magnetic fields 255]. The magnetic properties of the NV spin structure of up to 100 pT above the skin [266]. Meanwhile, fields enable magneto-optical nanoscopy, wherein fluorescent due to neural activity deep within the temporal lobe of nanodiamonds can be utilized as photostable tracers in the brain are typically below the 10 fT level at the scalp, biological systems, tracking intercellular protein trans- and require a high number of sensors to accurately re- port [256], or as magnetic tracers localized in a magnetic construct the source location [267]. field gradient, in analogy to MRI technique [257]. Significant progress has been made in mitigating the challenges associated with nanodiamond based magne- C. Magnetometry of the brain, nerves and muscles tometry (as opposed to bulk), including random orien- tations of the NV centers and broad thermal distribu- Despite the challenges, magnetometry of brain func- tions of nanodiamond ensembles [258, 259]. Opening up tion, or magnetoencephalography (MEG), is a powerful further applications, fluorescent nanodiamonds were re- tool with high spatial and temporal resolution, providing cently combined with biomimetic substrates used in the a direct measure of the electrical activity of neurons, in laboratory. ODMR spectra were obtained from live dif- contrast to other imaging techniques such as fMRI, PET ferentiated neural stem cells functioning as a connected or EEG. Furthermore, MEG is completely noninvasive, neural network grown on fluorescent nanodiamond em- making it an attractive option for both medical and tech- bedded nanofibers [260]. See [255] and references therein nological applications such as brain-computer interfaces for a detailed review on the use of nanodiamonds in cells. (BCI) [268]. Vapor cell OPMs have generally been considered too Clinically relevant measurements of magnetic fields large in size to have the resolution necessary for mag- from brain activity require both high sensitivity to de- netic detection in cellular biology. Recent advances in tect the fields and high sensor density to perform accu- chip-scale vapor cell fabrication have made possible cm- rate source localization [269]. Until recently, only SQUID size OPMs based on mm-scale cells (Sec. II), but below technologies were mature enough to meet these require- these scales, spin-destructive collisions with the cell walls ments, generating a global MEG industry. SQUID MEG dominate the spin relaxation rates, creating a trade-off systems mitigate the environmental challenges of detect- between vapor cell size and sensitivity [44, 261]. One ap- ing brain activity in a hospital by operating in magneti- proach around the larger vapor-cell size is the use of flux cally shielded environments and using gradiometric sen- guides, as described in Sec. III A. sors. Typically, these devices incorporate hundreds of sensors into a rigid head enclosure that is connected to a dewar that provides the necessary cryogenic cooling. B. Inside the human body While the rigidity and fixed position of the enclosure is ideal for magnetometer operation and source localization, Virtually all processes within the human body are con- incorporating sensitive magnetometry into natural envi- trolled and regulated via electrochemical signals. Non- ronments for subjects, ideally without magnetic shield- contact measurements of magnetic fields associated with ing, remains an open challenge. nerve and muscle activity are of great interest because Widely considered, therefore, are OPMs and NV mag- these signals can communicate detailed temporal and netometers, which, due to their small size, can move spatial information [262]. This information is highly com- freely with the user, untethered by cryogenic cooling. plementary to existing techniques to measure electromag- While NV magnetometers are rapidly approaching sen- netic activity within the human body, such as electroen- sitivities necessary to measure magnetic fields from the cephalography (EEG) or electromyography (EMG), as brain [252], OPMs have already overcome this thresh- well as to those techniques measuring the dynamics of old. Recent demonstrations with OPMs have shown that blood flow (hemodynamics) associated with neural ac- they can be placed closer to the scalp than comparable tivity, such as functional magnetic resonance imaging SQUID sensors, where the benefit is not only in yield- (fMRI), positron-emission tomography (PET), functional ing higher signal amplitudes from neurological magnetic near-infrared spectroscopy (fNIRS), and single-photon sources, but also in improved spatial information density, emission computed tomography (SPECT). From a medi- which outweighs benefits of higher sensitivity or of more 18 sensors [269–272]. To this end, multichannel magnetom- of oscillating magnetic fields can be used to discrimi- etry with OPMs within millimeters of the scalp has been nate different tissue types in the context of conductivity studied with up to 25 magnetometers [273, 274], includ- measurements or low-field NMR sensing. In particular, ing demonstrations of wearable systems [275]. discrimination between healthy and tumorous tissue us- Yet movement of these sensors places stringent require- ing local conductivity measurement [280, 281], appears ments on the background field. A subject may move their promising and might be within reach of optimized labora- head within a 30 cm cubed region, which in turn must tory OPMs [205]. In the field of functional neuroimaging be homogeneous to a high degree. Gradients as small via MEG (cf. V C), optically pumped alkali vapor-cell as 1 nT/cm could cause large artifacts that can either magnetometers are already employed in new modalities swamp or be confused with brain signals. As a result, due to their reduced sensor-to-skull distance and small sensor tracking (for example optically, as mentioned in overall sensor size. Going a step further and providing Sec. III E) within well characterized field regions, and ac- magnetic information from inside the body using needle tive shielding where the background-field noise is actively magnetometers could dramatically improve source loca- zeroed using the magnetometer signal [276, 277], can of- tion, especially in conjunction with a dense sensor net- fer benefits over passive shielding (Sec. III B). work around the skull. An important challenge in MEG is in source localiza- tion, which requires solving an ill-posed inverse problem. Well-defined head shape (using, for example, MRI of the D. Plant magnetometry subject) can constrain the solutions, but also crucial is sufficiently dense field mapping with vector information. Members of the plant kingdom have been probed in OPMs and NVs can be readily used as vector sensors exquisite detail using a variety of sensing modalities. Yet in at least two dimensions, and NVs are particularly literature covering magnetic sensing of plant processes is well suited to high densities of individual magnetome- limited [282], perhaps owing to a) the extremely small ters. While OPMs are limited by vapor-cell size, rel- expected signal size, b) difficulties in developing reliable atively high sensor densities can still be achieved. At stimulation protocols in the plant, and c) the relatively these sensor densities, however, cross-talk between sen- long timescale of plant signalling (1-5 s) - a regime of sors and miniaturization with heating could be problem- abundant low-frequency noise. Since the plant action po- atic. Cross-talk can be addressed via prior characteri- tential amplitudes are between 5-10 times smaller than zation, compensation and shielding, or all-optical imple- in animals and the propagation speeds are typically four mentations (Sec II A). orders of magnitude smaller [283], there are grave impli- Magnetic detection of neural and muscle activity of cations for the accompanying magnetic field amplitudes other body parts is also of clinical interest. While generated by most plants, again strictly limiting the mag- facing similar implementation challenges as magnetoen- netometer types that could be used for detecting. cephalography, magnetic signals from muscles are on the Nevertheless, successful attempts have been made, in order of 1 pT near the surface of the skin. Measurements particular with the use of SQUID magnetometers to mea- of muscle signals, or magnetomyography, have been per- sure signalling in wounded bean plants [284], and OPMs formed with the sensors described above, and, in com- to measure stimulated action potentials in Venus Fly- bination with other technologies such as TMS, can of- traps [285]. fer tools to study the human motor system with high spatio-temporal resolution (see Sec. III B). While these relatively large muscle signals are useful when the object VI. CONCLUDING REMARKS of interest is the muscle itself, they can obscure mag- netic signals that accompany nerve transmission, which As illustrated in this review, OPM’s have reached the are also of medical interest [278]. Internal magnetome- operational sensitivity to enable a wide range of appli- try with, for example, magnetic endoscopes, offer higher cations. It is the hope of the authors that this review SNR (due to the reduced distance to sources) and spa- would provide the reader with a set of general and spe- tial filtering, with added challenges of biocompatibility cific ideas of how to meet various diverse challenges aris- and further miniaturization. Needle-sized magnetome- ing in the way of real-life precision magnetic (or other) ters, previously implemented with giant magnetoresis- measurements. Extrapolating from the rapid progress tance sensors [279], are possible with diamond OPMs and made over the past decade, we can expect a future that could measure previously inaccessible in-vivo fields. This includes not just higher sensitivity but further miniatur- could enable new diagnostic tools in several fields, for ex- ization (with compact shielding), resiliency to harsher en- ample, cardiology, functional neuro-imaging and brain vironments, more extensive networks enabled by modern surgery. In the brain (or in the peripheral nerves or computational techniques, and further integration into the spine), magnetometers could measure neural activ- multi-functional hybrid sensors, all of which will open ity with sub-mm resolution, which would allow, given the application space further. the appropriate sensitivity, functional monitoring of in- There are also many open questions. While this review dividual nerves or structures. Additionally, detection mostly focuses on passive sensing devices, self-oscillating 19 magnetometers [228], lasers operating on magnetically ACKNOWLEDGEMENTS sensitive transitions, and spin masers [286–288] belong to the category of “active” sensing devices. The spectral The authors are grateful to L. Bougas, I. Fescenko, M. width of the signal generated by active devices could be Jiang, D. C. Hovde, J. Hruby, P. Kehayias, K. Kirch, T. orders of magnitude narrower than the width of the un- Kornack, P. Koss, R. Liu, P. Maletinsky, S. Prawer, M. derlying atomic line. It is thus tempting to ask: may such Romalis, T. Scholtes, S. Ulmer, Y. Semertzidis, and R. devices offer fundamental advantages over their “passive” Zhang for useful discussions. This work was supported by counterparts? In all numerous specific cases known to the EU FET-OPEN Flagship Project ASTERIQS (Ac- us, the answer is no; however, we are not aware of a tion 820394) and the German Federal Ministry of Educa- general proof that this should be the case. Even if ac- tion and Research (BMBF) within the Quantumtechnolo- tive devices cannot fundamentally outperform their pas- gien program (FKZ 13N14439 and FKZ 13N15064). DB sive counterparts, they are already proven to often be was supported in part by the DFG Project ID 390831469: advantageous in practice, for example with moving plat- EXC 2118 (PRISMA+ Cluster of Excellence). DB also forms (Sec. IV H) and for detecting low-frequency mag- received support from the European Research Council netic fields. Additional open questions include the rela- (ERC) under the European Union Horizon 2020 Research tive fundamental and practical performance of intrinsic and Innovation Program (grant agreement No. 695405) vs. composite gradiometers (Sec. III D 1), whether solid- and from the DFG Reinhart Koselleck Project. KMCF state OPMs will reach sensitivities comparable to those recognizes support from the HIM Visiting Scientist Pro- of vapor-cell OPMs, and if there are far superior solid gram and the National Science Foundation under Grant state defects, relative to the NV center, for sensing ap- No. CHE-1607869. plications [289, 290]. Finally, this review would be incomplete without men- tioning the high-precision Lorentz-invariance tests car- DATA AVAILABILITY ried out with atomic sensors at the South Pole [291], as an example of a challenging measurement in more than Data sharing is not applicable to this article as no new one sense. data were created or analyzed in this study.

[1] M. V. Balabas, T. Karaulanov, M. P. Ledbetter, and [13] T. Zhang, G.-Q. Liu, W.-H. Leong, C.-F. Liu, M.-H. D. Budker, Phys. Rev. Lett. 105, 070801 (2010). Kwok, T. Ngai, R.-B. Liu, and Q. Li, Nature Commu- [2] D. Budker and D. F. J. Kimball, Optical magnetometry nications 9, 3188 (2018). (Cambridge University Press, 2013). [14] N. T. Son, C. P. Anderson, A. Bourassa, K. C. Miao, [3] H. B. Dang, A. C. Maloof, and M. V. Romalis, Applied C. Babin, M. Widmann, M. Niethammer, J. Ul Hassan, Physics Letters 97, 151110 (2010). N. Morioka, I. G. Ivanov, F. Kaiser, J. Wrachtrup, and [4] J.-H. Storm, P. H¨ommen,D. Drung, and R. K¨orber, D. D. Awschalom, Applied Physics Letters 116, 190501 Applied Physics Letters 110, 072603 (2017). (2020). [5] M. Schmelz, V. Zakosarenko, A. Chwala, T. Sch¨onau, [15] D. Simin, V. A. Soltamov, A. V. Poshakinskiy, A. N. R. Stolz, S. Anders, S. Linzen, and H. Meyer, Anisimov, R. A. Babunts, D. O. Tolmachev, E. N. IEEE Transactions on Applied Superconductivity 26, Mokhov, M. Trupke, S. A. Tarasenko, A. Sperlich, P. G. 1 (2016). Baranov, V. Dyakonov, and G. V. Astakhov, Phys. Rev. [6] J. F. Barry, J. M. Schloss, E. Bauch, M. J. Turner, C. A. X 6, 031014 (2016). Hart, L. M. Pham, and R. L. Walsworth, Rev. Mod. [16] H. Kraus, V. A. Soltamov, F. Fuchs, D. Simin, A. Sper- Phys. 92, 015004 (2020). lich, P. G. Baranov, G. V. Astakhov, and V. Dyakonov, [7] E. Schaefer-Nolte, F. Reinhard, M. Ternes, Scientific Reports 4, 5303 (2014). J. Wrachtrup, and K. Kern, Review of Scientific [17] D. J. Christle, A. L. Falk, P. Andrich, P. V. Klimov, Instruments 85, 013701 (2014). J. U. Hassan, N. T. Son, E. Janz´en,T. Ohshima, and [8] D. M. Toyli, D. J. Christle, A. Alkauskas, B. B. Buckley, D. D. Awschalom, Nature Materials 14, 160 (2015). C. G. Van de Walle, and D. D. Awschalom, Phys. Rev. [18] M. Niethammer, M. Widmann, S.-Y. Lee, P. Stenberg, X 2, 031001 (2012). O. Kordina, T. Ohshima, N. T. Son, E. Janz´en, and [9] M. W. Doherty, V. V. Struzhkin, D. A. Simpson, L. P. J. Wrachtrup, Phys. Rev. Applied 6, 034001 (2016). McGuinness, Y. Meng, A. Stacey, T. J. Karle, R. J. [19] C. Abel, S. Afach, N. Ayres, G. Ban, G. Bison, Hemley, N. B. Manson, L. C. L. Hollenberg, and K. Bodek, V. Bondar, E. Chanel, P.-J. Chiu, C. Craw- S. Prawer, Phys. Rev. Lett. 112, 047601 (2014). ford, et al., arXiv preprint arXiv:1912.04631 (2019). [10] H. Mamin, M. Kim, M. Sherwood, C. Rettner, K. Ohno, [20] W. E. Bell and A. L. Bloom, Phys. Rev. Lett. 6, 280 D. Awschalom, and D. Rugar, Science 339, 557 (2013). (1961). [11] P. Maletinsky, S. Hong, M. S. Grinolds, B. Hausmann, [21] E. B. Alexandrov, M. Auzinsh, D. Budker, D. F. Kim- M. D. Lukin, R. L. Walsworth, M. Loncar, and A. Ya- ball, S. M. Rochester, and V. V. Yashchuk, J. Opt. Soc. coby, Nature Nanotechnology 7, 320 (2012). Am. B 22, 7 (2005). [12] C. L. Degen, Applied Physics Letters 92, 243111 (2008). 20

[22] Z. D. Gruji´c,P. A. Koss, G. Bison, and A. Weis, The [44] Y. J. Kim and I. Savukov, Scientific Reports 6, 24773 European Physical Journal D 69, 135 (2015). (2016). [23] B. Patton, E. Zhivun, D. Hovde, and D. Budker, Phys- [45] I. Savukov and T. Karaulanov, Journal of Magnetic Res- ical review letters 113, 013001 (2014). onance 249, 49 (2014). [24] E. Zhivun, A. Wickenbrock, B. Patton, and D. Budker, [46] F. Thiel, A. Schnabel, S. Knappe-Gr¨uneberg, D. Stoll- Applied Physics Letters 105, 192406 (2014). fuß, and M. Burghoff, Review of Scientific Instruments [25] H. Zheng, G. Chatzidrosos, A. Wickenbrock, L. Bougas, 78, 035106 (2007). R. Lazda, A. Berzins, F. H. Gahbauer, M. Auzinsh, [47] Z. Sun, P. Fierlinger, J. Han, L. Li, T. Liu, A. Schn- R. Ferber, and D. Budker, in Slow Light, Fast Light, abel, S. Stuiber, and J. Voigt, IEEE Transactions on and Opto-Atomic Precision Metrology X , Vol. 10119, Industrial Electronics , 1 (2020). edited by S. M. Shahriar and J. Scheuer, International [48] V. Yashchuk, S. Lee, and E. Paperno, “Magnetic shield- Society for Optics and Photonics (SPIE, 2017) pp. 115 ing,” in Optical Magnetometry, edited by D. Budker and – 122. J. D (Cambridge University Press, 2013) p. 225. [26] M. Auzinsh, A. Berzins, D. Budker, L. Busaite, R. Fer- [49] R. Hari and R. Salmelin, NeuroImage 61, 386 (2012). ber, F. Gahbauer, R. Lazda, A. Wickenbrock, and [50] G. Z. Iwata, Y. Hu, T. Sander, M. Muthuraman, V. C. H. Zheng, Phys. Rev. B 100, 075204 (2019). Chirumamilla, S. Groppa, D. Budker, and A. Wicken- [27] A. Wickenbrock, H. Zheng, L. Bougas, N. Leefer, brock, arXiv preprint arXiv:1909.11451 (2019). S. Afach, A. Jarmola, V. M. Acosta, and D. Budker, [51] P. J. Broser, S. Knappe, D.-S. Kajal, N. Noury, O. Alem, Applied Physics Letters 109, 053505 (2016). V. Shah, and C. Braun, IEEE Transactions on Neu- [28] V. M. Acosta, K. Jensen, C. Santori, D. Budker, and ral Systems and Rehabilitation Engineering 26, 2226 R. G. Beausoleil, Physical review letters 110, 213605 (2018). (2013). [52] M. R. Keshtkaran and Z. Yang, Journal of neural engi- [29] R. Akhmedzhanov, L. Gushchin, N. Nizov, V. Nizov, neering 11, 026017 (2014). D. Sobgayda, I. Zelensky, and P. Hemmer, Phys. Rev. [53] B. Widrow, J. R. Glover, J. M. McCool, J. Kaunitz, A 96, 013806 (2017). C. S. Williams, R. H. Hearn, J. R. Zeidler, J. Eugene [30] H. Zheng, Z. Sun, G. Chatzidrosos, C. Zhang, K. Naka- Dong, and R. C. Goodlin, Proceedings of the IEEE 63, mura, H. Sumiya, T. Ohshima, J. Isoya, J. Wrachtrup, 1692 (1975). A. Wickenbrock, and D. Budker, Phys. Rev. Applied [54] S. Ikeda and K. Toyama, Neural Networks 13, 1063 13, 044023 (2020). (2000). [31] M. Block, B. Kobrin, A. Jarmola, S. Hsieh, C. Zu, [55] T. Wolf, P. Neumann, K. Nakamura, H. Sumiya, N. Figueroa, V. Acosta, J. Minguzzi, J. Maze, D. Bud- T. Ohshima, J. Isoya, and J. Wrachtrup, Phys. Rev. ker, et al., arXiv preprint arXiv:2004.02886 (2020). X 5, 041001 (2015). [32] C. Harrison and J. Southam, Journal of geomagnetism [56] M. Tseytlin, Zeitschrift fur physikalische Chemie and geoelectricity 43, 585 (1991). (Frankfurt am Main, Germany) 231, 689 (2017). [33] P. Leroy, C. Coillot, V. Mosser, A. Roux, and [57] T. W. Parks and C. S. Burrus, Digital filter design G. Chanteur, Sensor Lett. 2, 162 (2007). (Wiley-Interscience, 1987). [34] A. S. Edelstein, G. A. Fischer, M. Pedersen, E. R. [58] A. Widmann and E. Schr¨oger,Frontiers in psychology Nowak, S. F. Cheng, and C. A. Nordman, Journal of 3, 233 (2012). Applied Physics 99, 08B317 (2006). [59] H. Failache, P. Valente, G. Ban, V. Lorent, and [35] M. J. Caruso, T. Bratland, C. H. Smith, and R. Schnei- A. Lezama, Phys. Rev. A 67, 043810 (2003). der, Sensors-Peterborough- 15, 34 (1998). [60] A. A. Wood, E. Lilette, Y. Y. Fein, N. Tomek, L. P. [36] R. C. Chaves, P. P. Freitas, B. Ocker, and W. Maass, McGuinness, L. C. L. Hollenberg, R. E. Scholten, and Journal of Applied Physics 103, 07E931 (2008). A. M. Martin, Science Advances 4, eaar7691 (2018). [37] S. Bondarenko, A. Shablo, P. Pavlov, and S. Perepelkin, [61] M. V. Romalis and H. B. Dang, Materials Today 14, Physica C: Superconductivity 372-376, 158 (2002). 258 (2011). [38] W. C. Griffith, R. Jimenez-Martinez, V. Shah, [62] Y.-X. Liu, A. Ajoy, and P. Cappellaro, Phys. Rev. Lett. S. Knappe, and J. Kitching, Applied Physics Letters 122, 100501 (2019). 94, 023502 (2009). [63] L. Viola, E. Knill, and S. Lloyd, Phys. Rev. Lett. 82, [39] I. Fescenko, A. Jarmola, I. Savukov, P. Kehayias, 2417 (1999). J. Smits, J. Damron, N. Ristoff, N. Mosavian, and [64] M. J. Biercuk, A. C. Doherty, and H. Uys, Journal of V. M. Acosta, Phys. Rev. Research 2, 023394 (2020). Physics B: Atomic, Molecular and Optical Physics 44, [40] M. Pannetier-Lecoeur, C. Fermon, N. Biziere, J. Scola, 154002 (2011). and A. L. Walliang, IEEE Transactions on Applied Su- [65] S. Kotler, N. Akerman, Y. Glickman, A. Keselman, and perconductivity 17, 598 (2007). R. Ozeri, Nature 473, 61 (2011). [41] Y. Xie, H. Yu, Y. Zhu, X. Qin, X. Rong, C.- [66] S. Upadhyay, U. Dargyte, D. Patterson, and J. D. We- K. Duan, and J. Du, Science Bulletin (2020), instein, Phys. Rev. Lett. 125, 043601 (2020). https://doi.org/10.1016/j.scib.2020.08.001. [67] A. A. Wood, A. G. Aeppli, E. Lilette, Y. Y. Fein, [42] A. Qasimi, C. Dolabdjian, D. Bloyet, and V. Mosser, A. Stacey, L. C. L. Hollenberg, R. E. Scholten, and IEEE Sensors Journal 4, 160 (2004). A. M. Martin, Phys. Rev. B 98, 174114 (2018). [43] A. S. Edelstein, G. Fischer, J. Pulskamp, M. Pedersen, [68] A. A. Wood, L. C. L. Hollenberg, R. E. Scholten, and W. Bernard, and S. F. Cheng, “Minimizing 1/f noise A. M. Martin, Phys. Rev. Lett. 124, 020401 (2020). in magnetic sensors with a mems flux concentrator,” in [69] A. V. Veryaskin, Gravity, Magnetic and Electromagnetic Transformational Science and Technology for the Cur- Gradiometry: Strategic Technologies in the 21st Century rent and Future Force (2006) pp. 24–29. (Morgan & Claypool Publishers, 2018). 21

[70] R. Zhang, R. Mhaskar, K. Smith, and M. Prouty, Ap- [88] C. Munuera-Javaloy, I. Arrazola, E. Solano, and plied Physics Letters 116, 143501 (2020). J. Casanova, Phys. Rev. B 101, 104411 (2020). [71] G. Oelsner, R. IJsselsteijn, T. Scholtes, A. Kr¨uger, [89] E. Bauch, C. A. Hart, J. M. Schloss, M. J. Turner, J. F. V. Schultze, G. Seyffert, G. Werner, M. J¨ager, Barry, P. Kehayias, S. Singh, and R. L. Walsworth, A. Chwala, and R. Stolz, arXiv preprint Phys. Rev. X 8, 031025 (2018). arXiv:2008.01570 (2020). [90] Z. Kazi, I. M. Shelby, H. Watanabe, K. M. Itoh, V. Shut- [72] M. Limes, E. Foley, T. Kornack, S. Caliga, S. McBride, thanandan, P. A. Wiggins, and K.-M. C. Fu, arXiv A. Braun, W. Lee, V. Lucivero, and M. Romalis, Phys- preprint arXiv:2002.06237 (2020). ical Review Applied 14, 011002 (2020). [91] B. J. Venema, P. K. Majumder, S. K. Lamoreaux, B. R. [73] V. Lucivero, W. Lee, N. Dural, and M. Romalis, arXiv Heckel, and E. N. Fortson, Phys. Rev. Lett. 68, 135 preprint arXiv:2009.13212 (2020). (1992). [74] T. M. Tierney, N. Holmes, S. Mellor, J. D. L´opez, [92] G. Vasilakis, J. M. Brown, T. W. Kornack, and M. V. G. Roberts, R. M. Hill, E. Boto, J. Leggett, V. Shah, Romalis, Phys. Rev. Lett. 103, 261801 (2009). M. J. Brookes, R. Bowtell, and G. R. Barnes, NeuroIm- [93] M. Bulatowicz, R. Griffith, M. Larsen, J. Mirijanian, age 199, 598 (2019). C. B. Fu, E. Smith, W. M. Snow, H. Yan, and T. G. [75] V. Angelopoulos, Space Science Reviews 141, 5 (2008). Walker, Phys. Rev. Lett. 111, 102001 (2013). [76] S. Afach, D. Budker, G. DeCamp, V. Dumont, [94] K. Tullney, F. Allmendinger, M. Burghoff, W. Heil, Z. Gruji´c,H. Guo, D. J. Kimball, T. Kornack, V. Lebe- S. Karpuk, W. Kilian, S. Knappe-Gr¨uneberg, dev, W. Li, H. Masia-Roig, S. Nix, M. Padniuk, W. M¨uller, U. Schmidt, A. Schnabel, F. Seifert, C. Palm, C. Pankow, A. Penaflor, X. Peng, S. Pustelny, Y. Sobolev, and L. Trahms, Phys. Rev. Lett. 111, T. Scholtes, J. Smiga, J. Stalnaker, A. Weis, A. Wick- 100801 (2013). enbrock, and D. Wurm, Physics of the Dark Universe [95] L. Hunter, J. Gordon, S. Peck, D. Ang, and J.-F. Lin, 22, 162 (2018). Science 339, 928 (2013). [77] H. Masia-Roig, J. A. Smiga, D. Budker, V. Dumont, [96] S. K. Lamoreaux, J. P. Jacobs, B. R. Heckel, F. J. Raab, Z. Grujic, D. Kim, D. F. J. Kimball], V. Lebedev, and E. N. Fortson, Phys. Rev. Lett. 58, 746 (1987). M. Monroy, S. Pustelny, T. Scholtes, P. C. Segura, [97] M. E. Limes, D. Sheng, and M. V. Romalis, Phys. Rev. Y. K. Semertzidis, Y. C. Shin, J. E. Stalnaker, I. Su- Lett. 120, 033401 (2018). lai, A. Weis, and A. Wickenbrock, Physics of the Dark [98] E. B. Alexandrov, M. V. Balabas, A. K. Vershovski, Universe 28, 100494 (2020). and A. S. Pazgalev, Technical Physics 49, 779 (2004). [78] F. Dolde, H. Fedder, M. W. Doherty, T. N¨obauer, [99] D. F. J. Kimball, I. Lacey, J. Valdez, J. Swiatlowski, F. Rempp, G. Balasubramanian, T. Wolf, F. Reinhard, C. Rios, R. Peregrina-Ramirez, C. Montcrieffe, J. Kre- L. C. L. Hollenberg, F. Jelezko, and J. Wrachtrup, Na- mer, J. Dudley, and C. Sanchez, Annalen der Physik ture Physics 7, 459 (2011). 525, 514 (2013). [79] T. Mittiga, S. Hsieh, C. Zu, B. Kobrin, F. Machado, [100] P. Gomez, F. Martin, C. Mazzinghi, D. Bene- P. Bhattacharyya, N. Z. Rui, A. Jarmola, S. Choi, dicto Orenes, S. Palacios, and M. W. Mitchell, Phys. D. Budker, and N. Y. Yao, Phys. Rev. Lett. 121, 246402 Rev. Lett. 124, 170401 (2020). (2018). [101] Z. Wang, X. Peng, R. Zhang, H. Luo, J. Li, Z. Xiong, [80] V. M. Acosta, E. Bauch, M. P. Ledbetter, A. Waxman, S. Wang, and H. Guo, Phys. Rev. Lett. 124, 193002 L.-S. Bouchard, and D. Budker, Phys. Rev. Lett. 104, (2020). 070801 (2010). [102] M. Ledbetter, S. Pustelny, D. Budker, M. Romalis, [81] D. A. Broadway, B. C. Johnson, M. S. J. Barson, S. E. J. Blanchard, and A. Pines, Physical review letters 108, Lillie, N. Dontschuk, D. J. McCloskey, A. Tsai, T. Ter- 243001 (2012). aji, D. A. Simpson, A. Stacey, J. C. McCallum, J. E. [103] T. Wu, J. W. Blanchard, D. F. Jackson Kimball, Bradby, M. W. Doherty, L. C. L. Hollenberg, and J. P. M. Jiang, and D. Budker, Phys. Rev. Lett. 121, 023202 Tetienne, Nano Letters, Nano Letters 19, 4543 (2019). (2018). [82] I. Fescenko, A. Laraoui, J. Smits, N. Mosavian, P. Ke- [104] S. Lourette, L. Bougas, M. Kayci, S. Xu, and D. Bud- hayias, J. Seto, L. Bougas, A. Jarmola, and V. M. ker, Journal of Applied Physics 126, 194502 (2019). Acosta, Phys. Rev. Applied 11, 034029 (2019). [105] M. Kayci, S. Lourette, T. Wang, L. Bougas, and [83] G. Kucsko, P. C. Maurer, N. Y. Yao, M. Kubo, H. J. D. Budker, Applied Physics Letters 113, 053103 (2018). Noh, P. K. Lo, H. Park, and M. D. Lukin, Nature [106] T. Lenz et. al., “Probing topological spin structures Physics Letter 500, 54 (2013). using concurrent light-polarization and magnetic mi- [84] D. M. Toyli, C. F. de las Casas, D. J. Christle, V. V. croscopy,” (2020), in preparation. Dobrovitski, and D. D. Awschalom, Proceedings of the [107] R. M. Hill, E. Boto, N. Holmes, C. Hartley, Z. A. See- National Academy of Sciences 110, 8417 (2013). dat, J. Leggett, G. Roberts, V. Shah, T. M. Tierney, [85] P. Neumann, I. Jakobi, F. Dolde, C. Burk, R. Reuter, M. W. Woolrich, C. J. Stagg, G. R. Barnes, R. Bowtell, G. Waldherr, J. Honert, T. Wolf, A. Brunner, J. H. R. Slater, and M. J. Brookes, Nature Communications Shim, D. Suter, H. Sumiya, J. Isoya, and J. Wrachtrup, 10, 4785 (2019). Nano Letters, Nano Letters 13, 2738 (2013). [108] N. Wang, G.-Q. Liu, W.-H. Leong, H. Zeng, X. Feng, [86] K. Fang, V. M. Acosta, C. Santori, Z. Huang, K. M. S.-H. Li, F. Dolde, H. Fedder, J. Wrachtrup, X.-D. Cui, Itoh, H. Watanabe, S. Shikata, and R. G. Beausoleil, S. Yang, Q. Li, and R.-B. Liu, Phys. Rev. X 8, 011042 Phys. Rev. Lett. 110, 130802 (2013). (2018). [87] H. J. Mamin, M. H. Sherwood, M. Kim, C. T. Rettner, [109] M. W. Doherty, V. M. Acosta, A. Jarmola, M. S. Bar- K. Ohno, D. D. Awschalom, and D. Rugar, Phys. Rev. son, N. B. Manson, D. Budker, and L. C. Hollenberg, Lett. 113, 030803 (2014). Physical Review B 90, 041201 (2014). 22

[110] M. Jiao, X. Rong, H. Liang, Y.-F. Cai, and J. Du, Phys. [133] Y. Schlussel, T. Lenz, D. Rohner, Y. Bar-Haim, Rev. D 101, 115011 (2020). L. Bougas, D. Groswasser, M. Kieschnick, E. Rozen- [111] F. Casola, T. van der Sar, and A. Yacoby, Nature Re- berg, L. Thiel, A. Waxman, J. Meijer, P. Maletinsky, views Materials 3, 1 (2018). D. Budker, and R. Folman, Phys. Rev. Applied 10, [112] A. Jayaraman, Rev. Mod. Phys. 55, 65 (1983). 034032 (2018). [113] R. Jim´enez-Mart´ınezand S. Knappe, “Microfabricated [134] A. Waxman, Y. Schlussel, D. Groswasser, V. M. Acosta, optically-pumped magnetometers,” in High Sensitivity L.-S. Bouchard, D. Budker, and R. Folman, Phys. Rev. Magnetometers, edited by A. Grosz, M. J. Haji-Sheikh, B 89, 054509 (2014). and S. C. Mukhopadhyay (Springer International Pub- [135] S. Hong, M. S. Grinolds, L. M. Pham, D. Le Sage, lishing, Cham, 2017) pp. 523–551. L. Luan, R. L. Walsworth, and A. Yacoby, MRS bul- [114] M. P. Ledbetter, I. M. Savukov, D. Budker, V. Shah, letin 38, 155 (2013). S. Knappe, J. Kitching, D. Michalak, S. Xu, and [136] L. Thiel, D. Rohner, M. Ganzhorn, P. Appel, E. Neu, A. Pines, Proceedings of the National Academy of Sci- B. M¨uller,R. Kleiner, D. Koelle, and P. Maletinsky, ences of the United States of America 105, 2286 (2008). Nature Nanotechnology , 677 (2016). [115] D. Duan, V. K. Kavatamane, S. R. Arumugam, G. Ra- [137] L. Thiel, Z. Wang, M. A. Tschudin, D. Rohner, hane, G.-X. Du, Y.-K. Tzeng, H.-C. Chang, and G. Bal- I. Guti´errez-Lezama,N. Ubrig, M. Gibertini, E. Gian- asubramanian, Opt. Lett. 44, 2851 (2019). nini, A. F. Morpurgo, and P. Maletinsky, Science 364, [116] I. Savukov and S. J. Seltzer, “Spin-exchange-relaxation- 973 (2019). free (SERF) magnetometers,” in Optical Magnetometry, [138] P. Maletinsky, “Personal communication,” (2020). edited by D. Budker and D. F. Jackson Kimball (Cam- [139] R. M. Hazen and R. T. Downs, High-Temperature and bridge University Press, 2013) p. 85–103. High Pressure Crystal Chemistry, Vol. 41 (Walter de [117] P. John, N. Polwart, C. Troupe, and J. Wilson, Dia- Gruyter GmbH & Co KG, 2018). mond and Related Materials 11, 861 (2002). [140] X. Wang and K. Kamenev, Low Temperature Physics [118] T. Evans and P. James, Proc. R. Soc. Lond. A 277, 260 40, 735 (2014). (1964). [141] Q. Wei, C. McCammon, and S. A. Gilder, Geochem- [119] S. Chakravarthi, C. Moore, A. Opsvig, C. Pederson, istry, Geophysics, Geosystems 18, 4646 (2017). E. Hunt, A. Ivanov, I. Christen, S. Dunham, and K.- [142] D. J. Dunlop and O.¨ Ozdemir,¨ Rock magnetism: fun- M. C. Fu, Phys. Rev. Materials 4, 023402 (2020). damentals and frontiers, Vol. 3 (Cambridge university [120] A. Jarmola, V. M. Acosta, K. Jensen, S. Chemerisov, press, 2001). and D. Budker, Phys. Rev. Lett. 108, 197601 (2012). [143] N. Dubrovinskaia, L. Dubrovinsky, N. A. Solopova, [121] N. Bar-Gill, L. M. Pham, A. Jarmola, D. Budker, A. Abakumov, S. Turner, M. Hanfland, E. Bykova, and R. L. Walsworth, Nature Communications 4, 1743 M. Bykov, C. Prescher, V. B. Prakapenka, S. Petitgi- (2013). rard, I. Chuvashova, B. Gasharova, Y.-L. Mathis, P. Er- [122] A. Norambuena, E. Mu˜noz,H. T. Dinani, A. Jarmola, shov, I. Snigireva, and A. Snigirev, Science Advances 2 P. Maletinsky, D. Budker, and J. R. Maze, Phys. Rev. (2016), 10.1126/sciadv.1600341. B 97, 094304 (2018). [144] G. Giriat, W. Wang, J. P. Attfield, A. D. Huxley, and [123] M. Mr´ozek, D. Rudnicki, P. Kehayias, A. Jarmola, K. V. Kamenev, Review of Scientific Instruments 81, D. Budker, and W. Gawlik, EPJ Quantum Technol- 073905 (2010). ogy 2, 22 (2015). [145] R. A. Sadykov, N. S. Bezaeva, A. I. Kharkovskiy, P. Ro- [124] G.-Q. Liu, X. Feng, N. Wang, Q. Li, and R.-B. Liu, chette, J. Gattacceca, and V. I. Trukhin, Review of Nature Communications 10, 1344 (2019). Scientific Instruments 79, 115102 (2008). [125] T. Plakhotnik, M. W. Doherty, J. H. Cole, R. Chapman, [146] K. Kitagawa, N. Katayama, H. Gotou, T. Yagi, K. Oh- and N. B. Manson, Nano Letters, Nano Letters 14, 4989 gushi, T. Matsumoto, Y. Uwatoko, and M. Takigawa, (2014). Physical review letters 103, 257002 (2009). [126] S. I. Kanorsky, S. Lang, S. L¨ucke, S. B. Ross, T. W. [147] C. Lin, K. Shirer, J. Crocker, A. Dioguardi, M. Lawson, H¨ansch, and A. Weis, Phys. Rev. A 54, R1010 (1996). B. Bush, P. Klavins, and N. J. Curro, Physical Review [127] M. Arndt, S. Kanorsky, A. Weis, and T. H¨ansch, B 92, 155147 (2015). Physics Letters A 174, 298 (1993). [148] S. Lee, K. Luszczynski, R. E. Norberg, and M. S. Con- [128] A. Hatakeyama, K. Oe, K. Ota, S. Hara, J. Arai, radi, Review of Scientific Instruments 58, 415 (1987). T. Yabuzaki, and A. R. Young, Phys. Rev. Lett. 84, [149] M. Lesik, T. Plisson, L. Toraille, J. Renaud, F. Occelli, 1407 (2000). M. Schmidt, O. Salord, A. Delobbe, T. Debuisschert, [129] A. Sushkov and D. Budker, Physical Review A 77, L. Rondin, P. Loubeyre, and J.-F. Roch, Science 366, 042707 (2008). 1359 (2019). [130] K. R. Joshi, N. Nusran, M. A. Tanatar, K. Cho, S. L. [150] S. Hsieh, P. Bhattacharyya, C. Zu, T. Mittiga, T. J. Bud’ko, P. C. Canfield, R. M. Fernandes, A. Levchenko, Smart, F. Machado, B. Kobrin, T. O. H¨ohn,N. Z. Rui, and R. Prozorov, New Journal of Physics (2020). M. Kamrani, S. Chatterjee, S. Choi, M. Zaletel, V. V. [131] K. Joshi, N. Nusran, M. Tanatar, K. Cho, W. Meier, Struzhkin, J. E. Moore, V. I. Levitas, R. Jeanloz, and S. Bud’ko, P. Canfield, and R. Prozorov, Phys. Rev. N. Y. Yao, Science 366, 1349 (2019). Applied 11, 014035 (2019). [151] K. Y. Yip, K. O. Ho, K. Y. Yu, Y. Chen, W. Zhang, [132] V. M. Acosta, L. S. Bouchard, D. Budker, R. Folman, S. Kasahara, Y. Mizukami, T. Shibauchi, Y. Matsuda, T. Lenz, P. Maletinsky, D. Rohner, Y. Schlussel, and S. K. Goh, and S. Yang, Science 366, 1355 (2019). L. Thiel, Journal of Superconductivity and Novel Mag- [152] Y.-X. Shang, F. Hong, J.-H. Dai, Hui-Yu, Y.-N. Lu, E.- netism 32, 85 (2019). K. Liu, X.-H. Yu, G.-Q. Liu, and X.-Y. Pan, Chinese 23

Physics Letters 36, 086201 (2019). Curmi, M. Sennour, A. Thorel, M. B¨orsch, K. Aulen- [153] N. Aslam, M. Pfender, P. Neumann, R. Reuter, bacher, R. Erdmann, P. R. Hemmer, F. Jelezko, and A. Zappe, F. F´avaro de Oliveira, A. Denisenko, J. Wrachtrup, ACS Nano 3, 1959 (2009). H. Sumiya, S. Onoda, J. Isoya, and J. Wrachtrup, Sci- [171] B. R. Smith, D. W. Inglis, B. Sandnes, J. R. Rabeau, ence 357, 67 (2017). A. V. Zvyagin, D. Gruber, C. J. Noble, R. Vogel, [154] J. M. Boss, K. Chang, J. Armijo, K. Cujia, T. Rosskopf, E. Osawa,¯ and T. Plakhotnik, Small 5, 1649 (2009). J. R. Maze, and C. L. Degen, Phys. Rev. Lett. 116, [172] C. Bradac, T. Gaebel, N. Naidoo, M. J. Sellars, 197601 (2016). J. Twamley, L. J. Brown, A. S. Barnard, T. Plakhotnik, [155] S. J. DeVience, L. M. Pham, I. Lovchinsky, A. O. A. V. Zvyagin, and J. R. Rabeau, Nature Nanotechnol- Sushkov, N. Bar-Gill, C. Belthangady, F. Casola, ogy 5, 345 (2010). M. Corbett, H. Zhang, M. Lukin, H. Park, A. Yacoby, [173] T. Gaebel, C. Bradac, J. Chen, J. Say, L. Brown, and R. L. Walsworth, Nature Nanotechnology 10, 129 P. Hemmer, and J. Rabeau, Diamond and Related Ma- (2015). terials 21, 28 (2012). [156] I. Lovchinsky, A. O. Sushkov, E. Urbach, N. P. de Leon, [174] C. Bradac, T. Gaebel, C. I. Pakes, J. M. Say, A. V. S. Choi, K. De Greve, R. Evans, R. Gertner, E. Bersin, Zvyagin, and J. R. Rabeau, Small 9, 132 (2013). C. M¨uller,L. McGuinness, F. Jelezko, R. L. Walsworth, [175] M. Kaviani, P. De´ak,B. Aradi, T. Frauenheim, J.-P. H. Park, and M. D. Lukin, Science 351, 836 (2016). Chou, and A. Gali, Nano Letters 14, 4772 (2014). [157] L. G. Steele, M. Lawson, M. Onyszczak, B. T. Bush, [176] S. Sangtawesin, B. L. Dwyer, S. Srinivasan, J. J. Z. Mei, A. P. Dioguardi, J. King, A. Parker, A. Pines, Allred, L. V. H. Rodgers, K. De Greve, A. Stacey, S. T. Weir, W. Evans, K. Visbeck, Y. K. Vohra, N. Dontschuk, K. M. O’Donnell, D. Hu, D. A. Evans, and N. J. Curro, Applied Physics Letters 111, 221903 C. Jaye, D. A. Fischer, M. L. Markham, D. J. Twitchen, (2017). H. Park, M. D. Lukin, and N. P. de Leon, Phys. Rev. [158] M. Pelliccione, A. Jenkins, P. Ovartchaiyapong, X 9, 031052 (2019). C. Reetz, E. Emmanouilidou, N. Ni, and A. C. Bleszyn- [177] B. A. Myers, A. Das, M. C. Dartiailh, K. Ohno, D. D. ski Jayich, Nature Nanotechnology 11, 700 (2016). Awschalom, and A. C. Bleszynski Jayich, Phys. Rev. [159] R. Battesti, J. Beard, S. B¨oser,N. Bruyant, D. Budker, Lett. 113, 027602 (2014). S. A. Crooker, E. J. Daw, V. V. Flambaum, T. Inada, [178] P. G. Baranov, A. A. Soltamova, D. O. Tolmachev, I. G. Irastorza, F. Karbstein, D. L. Kim, M. G. Kozlov, N. G. Romanov, R. A. Babunts, F. M. Shakhov, S. V. Z. Melhem, A. Phipps, P. Pugnat, G. Rikken, C. Rizzo, Kidalov, A. Y. Vul’, G. V. Mamin, S. B. Orlinskii, and M. Schott, Y. K. Semertzidis, H. H. ten Kate, and N. I. Silkin, Small 7, 1533 (2011). G. Zavattini, Physics Reports 765-766, 1 (2018). [179] S. Sotoma, D. Terada, T. F. Segawa, R. Igarashi, [160] A. den Ouden, C. A. Wulffers, N. E. Hussey, G. Laureijs, Y. Harada, and M. Shirakawa, Scientific Reports 8, F. J. P. Wijnen, G. F. A. J. Wulterkens, M. D. Bird, I. R. 5463 (2018). Dixon, and J. A. A. J. Perenboom, IEEE Transactions [180] M. E. Trusheim, L. Li, A. Laraoui, E. H. Chen, on Applied Superconductivity 26, 1 (2016). H. Bakhru, T. Schr¨oder,O. Gaathon, C. A. Meriles, [161] A. Nikiel, P. Bl¨umler,W. Heil, M. Hehn, S. Karpuk, and D. Englund, Nano Letters 14, 32 (2014). A. Maul, E. Otten, L. M. Schreiber, and M. Terekhov, [181] J. Maze, P. Stanwix, J. Hodges, S. Hong, J. Taylor, The European Physical Journal D 68, 330 (2014). P. Cappellaro, L. Jiang, M. G. Dutt, E. Togan, A. Zi- [162] S. George, N. Bruyant, J. B´eard, S. Scotto, E. Ari- brov, et al., Nature 455, 644 (2008). mondo, R. Battesti, D. Ciampini, and C. Rizzo, Review [182] G. Balasubramanian, I. Chan, R. Kolesov, M. Al- of Scientific Instruments 88, 073102 (2017). Hmoud, J. Tisler, C. Shin, C. Kim, A. Wojcik, P. R. [163] D. Nakamura, H. Sawabe, Y. H. Matsuda, and Hemmer, A. Krueger, et al., Nature 455, 648 (2008). S. Takeyama, Review of Scientific Instruments 84, [183] K. Ohashi, T. Rosskopf, H. Watanabe, M. Loretz, 044702 (2013). Y. Tao, R. Hauert, S. Tomizawa, T. Ishikawa, J. Ishi- [164] M. S. Safronova, D. Budker, D. DeMille, D. F. J. Kim- Hayase, S. Shikata, C. L. Degen, and K. M. Itoh, Nano ball, A. Derevianko, and C. W. Clark, Rev. Mod. Phys. Letters 13, 4733 (2013). 90, 025008 (2018). [184] B. K. Ofori-Okai, S. Pezzagna, K. Chang, M. Loretz, [165] J. W. Blanchard and D. Budker, “Zero- to Ultralow- R. Schirhagl, Y. Tao, B. A. Moores, K. Groot-Berning, Field NMR,” in eMagRes (American Chemical Society, J. Meijer, and C. L. Degen, Phys. Rev. B 86, 081406 2016) pp. 1395–1410. (2012). [166] M. Jiang, R. P. Frutos, T. Wu, J. W. Blanchard, [185] Y. Dovzhenko, F. Casola, S. Schlotter, T. X. Zhou, X. Peng, and D. Budker, Phys. Rev. Applied 11, 024005 F. B¨uttner,R. L. Walsworth, G. S. D. Beach, and (2019). A. Yacoby, Nature Communications 9, 2712 (2018). [167] H. Zheng, J. Xu, G. Z. Iwata, T. Lenz, J. Michl, [186] I. Gross, W. Akhtar, V. Garcia, L. J. Mart´ınez, B. Yavkin, K. Nakamura, H. Sumiya, T. Ohshima, S. Chouaieb, K. Garcia, C. Carr´et´ero,A. Barth´el´emy, J. Isoya, J. Wrachtrup, A. Wickenbrock, and D. Bud- P. Appel, P. Maletinsky, J. V. Kim, J. Y. Chauleau, ker, Phys. Rev. Applied 11, 064068 (2019). N. Jaouen, M. Viret, M. Bibes, S. Fusil, and V. Jacques, [168] A. K. Vershovskii and A. K. Dmitriev, Technical Physics Nature 549, 252 (2017). 65, 1301 (2020). [187] P. Appel, M. Ganzhorn, E. Neu, and P. Maletinsky, [169] T. Lenz, A. Wickenbrock, F. Jelezko, G. Balasubrama- New Journal of Physics 17, 112001 (2015). nian, and D. Budker, arXiv preprint arXiv:2009.12117 [188] K. Chang, A. Eichler, J. Rhensius, L. Lorenzelli, and (2020). C. L. Degen, Nano Letters 17, 2367 (2017). [170] J. Tisler, G. Balasubramanian, B. Naydenov, [189] S. Ernst, D. M. Irber, A. M. Waeber, G. Braunbeck, R. Kolesov, B. Grotz, R. Reuter, J.-P. Boudou, P. A. and F. Reinhard, ACS Photonics 6, 327 (2019). 24

[190] A. Jenkins, M. Pelliccione, G. Yu, X. Ma, X. Li, K. L. [212] R. IJsselsteijn, M. Kielpinski, S. Woetzel, T. Scholtes, Wang, and A. C. B. Jayich, Phys. Rev. Materials 3, E. Kessler, R. Stolz, V. Schultze, and H.-G. Meyer, 083801 (2019). Review of Scientific Instruments 83, 113106 (2012). [191] P. Kehayias, A. Jarmola, N. Mosavian, I. Fescenko, [213] B. Patton, A. Brown, R. Slocum, and E. Smith, “Space F. M. Benito, A. Laraoui, J. Smits, L. Bougas, D. Bud- magnetometry,” in Optical Magnetometry, edited by ker, A. Neumann, S. R. J. Brueck, and V. M. Acosta, D. Budker and J. D (Cambridge University Press, 2013) Nature Communications 8, 188 (2017). p. 285. [192] S. A. Momenzadeh, R. J. St¨ohr, F. F. de Oliveira, [214] I. Mateos, B. Patton, E. Zhivun, D. Budker, D. Wurm, A. Brunner, A. Denisenko, S. Yang, F. Reinhard, and and J. Ramos-Castro, Sensors and Actuators A: Physi- J. Wrachtrup, Nano Letters 15, 165 (2015). cal 224, 147 (2015). [193] D. J. McCloskey, N. Dontschuk, D. A. Broadway, [215] J. Hruby, G. Magchiels, T. Mladenov, D. Gybels, A. Nadarajah, A. Stacey, J.-P. Tetienne, L. C. L. Hol- K. Winter, J. Vodnik, J. Prooth, M. Gulka, E. Bour- lenberg, S. Prawer, and D. A. Simpson, ACS Applied geois, and M. Nesladek, in 69th International Astro- Materials & Interfaces 12, 13421 (2020). nautical Congress, Bremen, Germany. IAC2018 (2018). [194] B. A. Myers, A. Ariyaratne, and A. C. B. Jayich, Phys. [216] D. M. Miles, I. R. Mann, M. Ciurzynski, D. Barona, Rev. Lett. 118, 197201 (2017). B. B. Narod, J. R. Bennest, I. P. Pakhotin, A. Kale, [195] M. Kim, H. J. Mamin, M. H. Sherwood, K. Ohno, B. Bruner, C. D. A. Nokes, C. Cupido, T. Haluza- D. D. Awschalom, and D. Rugar, Phys. Rev. Lett. 115, DeLay, D. G. Elliott, and D. K. Milling, Journal of Geo- 087602 (2015). physical Research: Space Physics 121, 11,839 (2016). [196] P. Appel, B. J. Shields, T. Kosub, N. Hedrich, [217] M. D´ıaz-Michelena, Sensors 9, 2271 (2009). R. H¨ubner, J. Faßbender, D. Makarov, and [218] J. M. Higbie, S. M. Rochester, B. Patton, R. Holzl¨ohner, P. Maletinsky, Nano Letters 19, 1682 (2019). D. B. Calia, and D. Budker, Proceedings of the National [197] J. Kitching, Applied Physics Reviews 5, 031302 (2018). Academy of Sciences 108, 3522 (2011). [198] F. Pedreros Bustos, D. Bonaccini Calia, D. Budker, [219] T. J. Kane, P. D. Hillman, C. A. Denman, M. Hart, M. Centrone, J. Hellemeier, P. Hickson, R. Holzl¨ohner, R. Phillip Scott, M. E. Purucker, and S. J. Potashnik, and S. Rochester, Nature Communications 9, 3981 Journal of Geophysical Research: Space Physics 123, (2018). 6171 (2018). [199] B. Zhou, Y. Yang, Y. Zhang, X. Gou, B. Cheng, [220] T. Fan, X. Yang, J. Dong, L. Zhang, S. Cui, J. Qian, J. Wang, and L. Li, Earth and Planetary Physics 2, R. Dong, K. Deng, T. Zhou, K. Wei, Y. Feng, and 455 (2018). W. Chen, Journal of Geophysical Research: Space [200] G. Chatzidrosos, A. Wickenbrock, L. Bougas, N. Leefer, Physics 124, 7505 (2019). T. Wu, K. Jensen, Y. Dumeige, and D. Budker, Phys. [221] B. Patton, O. Versolato, D. C. Hovde, E. Corsini, J. M. Rev. Applied 8, 044019 (2017). Higbie, and D. Budker, Applied Physics Letters 101, [201] A. Horsley, P. Appel, J. Wolters, J. Achard, A. Tallaire, 083502 (2012). P. Maletinsky, and P. Treutlein, Phys. Rev. Applied [222] E. E. McDonnell, S. Han, C. Hilty, K. L. Pierce, and 10, 044039 (2018). A. Pines, Analytical Chemistry 77, 8109 (2005). [202] Y. Mu, X. Zhang, W. Xie, and Y. Zheng, Remote Sens- [223] S. Xu, V. V. Yashchuk, M. H. Donaldson, S. M. ing 12, 452 (2020). Rochester, D. Budker, and A. Pines, Proceedings of [203] E. Klinger, H. Azizbekyan, A. Sargsyan, C. Leroy, the National Academy of Sciences 103, 12668 (2006). D. Sarkisyan, and A. Papoyan, Appl. Opt. 59, 2231 [224] S. Xu, E. Harel, D. J. Michalak, C. W. Crawford, (2020). D. Budker, and A. Pines, Journal of Magnetic Reso- [204] L. Bougas, L. D. Langenegger, C. A. Mora, M. Zeltner, nance Imaging 28, 1299 (2008). W. J. Stark, A. Wickenbrock, J. W. Blanchard, and [225] C. P. Gooneratne, B. Li, and T. E. Moellendick, Sensors D. Budker, Scientific Reports 8, 3491 (2018). 17, 2384 (2017). [205] G. Chatzidrosos, A. Wickenbrock, L. Bougas, H. Zheng, [226] M. I. Ibrahim, C. Foy, D. Kim, D. R. Englund, and O. Tretiak, Y. Yang, and D. Budker, Phys. Rev. Ap- R. Han, in 2018 IEEE Symposium on VLSI Circuits plied 11, 014060 (2019). (2018) pp. 249–250. [206] D. Le Sage, K. Arai, D. R. Glenn, S. J. DeVience, [227] H. Clevenson, L. M. Pham, C. Teale, K. Johnson, D. En- L. M. Pham, L. Rahn-Lee, M. D. Lukin, A. Yacoby, glund, and D. Braje, Applied Physics Letters 112, A. Komeili, and R. L. Walsworth, Nature 496, 486 252406 (2018), https://doi.org/10.1063/1.5034216. (2013). [228] J. Higbie, E. Corsini, and D. Budker, Review of scien- [207] A. Wickenbrock, S. Jurgilas, A. Dow, L. Marmugi, and tific instruments 77, 113106 (2006). F. Renzoni, Opt. Lett. 39, 6367 (2014). [229] G. Oelsner, V. Schultze, R. IJsselsteijn, F. Wittk¨amper, [208] A. Wickenbrock, N. Leefer, J. W. Blanchard, and and R. Stolz, Phys. Rev. A 99, 013420 (2019). D. Budker, Applied Physics Letters 108, 183507 (2016). [230] W. F. Stuart, Reports on Progress in Physics 35, 803 [209] L. Marmugi, C. Deans, and F. Renzoni, Applied Physics (1972). Letters 115, 083503 (2019). [231] M. N. Nabighian, M. E. Ander, V. J. S. Grauch, R. O. [210] K. Jensen, M. Zugenmaier, J. Arnbak, H. Stærkind, Hansen, T. R. LaFehr, Y. Li, W. C. Pearson, J. W. M. V. Balabas, and E. S. Polzik, Phys. Rev. Research Peirce, J. D. Phillips, and M. E. Ruder, in Geophysics 1, 033087 (2019). Today: A Survey of the Field as the Journal Celebrates [211] C. Deans, L. Marmugi, and F. Renzoni, Ap- its 75th Anniversary (Society of Exploration Geophysi- plied Physics Letters 116, 133501 (2020), cists, 2010). https://doi.org/10.1063/5.0002146. [232] S. J. Seltzer, P. J. Meares, and M. V. Romalis, Phys. Rev. A 75, 051407 (2007). 25

[233] V. Acosta, M. P. Ledbetter, S. M. Rochester, D. Bud- [245] S. Corrodi, P. De Lurgio, D. Flay, J. Grange, R. Hong, ker, D. F. Jackson Kimball, D. C. Hovde, W. Gawlik, D. Kawall, M. Oberling, S. Ramachandran, and P. Win- S. Pustelny, J. Zachorowski, and V. V. Yashchuk, Phys. ter, arXiv preprint arXiv:2003.06244 (2020). Rev. A 73, 053404 (2006). [246] M. Farooq, T. Chupp, J. Grange, A. Tewsley-Booth, [234] K. Jensen, V. M. Acosta, J. M. Higbie, M. P. Ledbetter, D. Flay, D. Kawall, N. Sachdeva, and P. Winter, Phys. S. M. Rochester, and D. Budker, Phys. Rev. A 79, Rev. Lett. 124, 223001 (2020). 023406 (2009). [247] J. A. Devlin, E. Wursten, J. A. Harrington, T. Higuchi, [235] G. Bao, A. Wickenbrock, S. Rochester, W. Zhang, and P. E. Blessing, M. J. Borchert, S. Erlewein, J. J. Hansen, D. Budker, Phys. Rev. Lett. 120, 033202 (2018). J. Morgner, M. A. Bohman, et al., Physical Review Ap- [236] A. Ben-Kish and M. V. Romalis, Phys. Rev. Lett. 105, plied 12, 044012 (2019). 193601 (2010). [248] S. Afach, G. Bison, K. Bodek, F. Burri, Z. Chowdhuri, [237] A. Chwala, M. Schmelz, V. Zakosarenko, M. Schif- M. Daum, M. Fertl, B. Franke, Z. Grujic, V. H´elaine, fler, M. Schneider, M. Th¨urk, S. Br¨auer, F. Bauer, R. Henneck, M. Kasprzak, K. Kirch, H.-C. Koch, M. Schulz, A. Kr¨uger, and R. Stolz, Superconductor A. Kozela, J. Krempel, B. Lauss, T. Lefort, Y. Lemi`ere, Science and Technology 32, 024003 (2019). M. Meier, O. Naviliat-Cuncic, F. M. Piegsa, G. Pignol, [238] J. D. Landry, F. C. Schoenig, L. N. Grossman, H. Berna- C. Plonka-Spehr, P. N. Prashanth, G. Qu´em´ener,D. Re- towicz, and I. S. Jacobs, Journal of Applied Physics 53, breyend, S. Roccia, P. Schmidt-Wellenburg, A. Schn- 8260 (1982). abel, N. Severijns, J. Voigt, A. Weis, G. Wyszynski, [239] I. Bolshakova, I. Duran, R. Holyaka, E. Hristoforou, J. Zejma, J. Zenner, and G. Zsigmond, Journal of Ap- and A. Marusenkov, Sensor Letters 5, 283 (2007). plied Physics 116, 084510 (2014). [240] D. C. Hovde, Millimeter Scale Magnetometer: Fi- [249] T. A. Bowen, E. Zhivun, A. Wickenbrock, V. Dumont, nal Technical Report, NASA contract 80NSSC18P2052, S. D. Bale, C. Pankow, G. Dobler, J. S. Wurtele, and Tech. Rep. (Southwest Sciences, 2019). D. Budker, Geoscientific Instrumentation, Methods and [241] B. Buffington, “Trajectory design concept for the pro- Data Systems 8, 129 (2019). posed Europa Clipper mission,” in AIAA/AAS Astro- [250] A. Kuwahata, T. Kitaizumi, K. Saichi, T. Sato, dynamics Specialist Conference (2014). R. Igarashi, T. Ohshima, Y. Masuyama, T. Iwasaki, [242] “Databases for radiation hardness,” M. Hatano, F. Jelezko, et al., Scientific Reports 10, 1 Https://lhcb-elec.web.cern.ch/lhcb- (2020). elec/html/radiation hardness.htm. [251] S. Steinert, F. Ziem, L. Hall, A. Zappe, M. Schweikert, [243] L. B¨ani, A. Alexopoulos, M. Artuso, F. Bach- N. G¨otz,A. Aird, G. Balasubramanian, L. Hollenberg, mair, M. Bartosik, J. Beacham, H. Beck, V. Bellini, and J. Wrachtrup, Nature communications 4, 1 (2013). V. Belyaev, B. Bentele, E. Berdermann, P. Bergonzo, [252] J. F. Barry, M. J. Turner, J. M. Schloss, D. R. Glenn, A. Bes, J.-M. Brom, M. Bruzzi, M. Cerv, G. Chio- Y. Song, M. D. Lukin, H. Park, and R. L. Walsworth, dini, D. Chren, V. Cindro, G. Claus, J. Collot, Proceedings of the National Academy of Sciences 113, J. Cumalat, A. Dabrowski, R. D'Alessandro, D. Dau- 14133 (2016). vergne, W. de Boer, C. Dorfer, M. D¨unser,V. Eremin, [253] G. Balasubramanian, A. Lazariev, S. R. Arumugam, R. Eusebi, G. Forcolin, J. Forneris, H. Frais-K¨olbl, and D.-w. Duan, Current opinion in chemical biology L. Gallin-Martel, M. Gallin-Martel, K. Gan, M. Gastal, 20, 69 (2014). C. Giroletti, M. Goffe, J. Goldstein, A. Golubev, [254] K. van der Laan, M. Hasani, T. Zheng, and A. Goriˇsek,E. Grigoriev, J. Grosse-Knetter, A. Grum- R. Schirhagl, Small 14, 1703838 (2018). mer, B. Gui, M. Guthoff, I. Haughton, B. Hiti, [255] M. Chipaux, K. J. van der Laan, S. R. Hemelaar, D. Hits, M. Hoeferkamp, T. Hofmann, J. Hosslet, M. Hasani, T. Zheng, and R. Schirhagl, Small 14, J.-Y. Hostachy, F. H¨ugging, C. Hutton, H. Jansen, 1704263 (2018). J. Janssen, H. Kagan, K. Kanxheri, G. Kasieczka, [256] Y. Kuo, T.-Y. Hsu, Y.-C. Wu, and H.-C. Chang, Bio- R. Kass, F. Kassel, M. Kis, V. Konovalov, G. Kram- materials 34, 8352 (2013). berger, S. Kuleshov, A. Lacoste, S. Lagomarsino, [257] B.-R. Lin, S. Kunuku, C.-H. Chen, T.-Y. Chen, T.-Y. A. L. Giudice, E. Lukosi, C. Maazouzi, I. Mandic, Hsiao, H.-K. Yu, Y.-J. Chang, L.-C. Liao, H. Niu, and C. Mathieu, M. Menichelli, M. Mikuˇz, A. Morozzi, C.-P. Lee, Scientific reports 9, 1 (2019). J. Moss, R. Mountain, S. Murphy, M. Muˇskinja, [258] A. M. Wojciechowski, P. Nakonieczna, M. Mr´ozek, A. Oh, P. Oliviero, D. Passeri, H. Pernegger, R. Per- K. Sycz, A. Kruk, M. Ficek, M. G lowacki, R. Bogdanow- rino, F. Picollo, M. Pomorski, R. Potenza, A. Quadt, icz, and W. Gawlik, Materials 12, 2951 (2019). A. Re, M. Reichmann, G. Riley, S. Roe, D. Sanz, [259] A. M. Wojciechowski, M. Karadas, C. Osterkamp, M. Scaringella, D. Schaefer, C. Schmidt, S. Schnetzer, S. Jankuhn, J. Meijer, F. Jelezko, A. Huck, and U. L. S. Sciortino, A. Scorzoni, S. Seidel, L. Servoli, S. Smith, Andersen, Applied Physics Letters 113, 013502 (2018). B. Sopko, V. Sopko, S. Spagnolo, S. Spanier, K. Sten- [260] J. C. Price, S. J. Levett, V. Radu, D. A. Simpson, A. M. son, R. Stone, C. Sutera, B. Tannenwald, A. Taylor, Barcons, C. F. Adams, and M. L. Mather, Small 15, M. Traeger, D. Tromson, W. Trischuk, C. Tuve, L. Up- 1900455 (2019). legger, J. Velthuis, N. Venturi, E. Vittone, S. Wagner, [261] J. Allred, R. Lyman, T. Kornack, and M. V. Romalis, R. Wallny, J. Wang, J. Weingarten, C. Weiss, T. Wen- Physical review letters 89, 130801 (2002). gler, N. Wermes, M. Yamouni, and M. Zavrtanik, Jour- [262] S. J. Williamson, G.-L. Romani, L. Kaufman, and nal of Instrumentation 13, C01029 (2018). I. Modena, Biomagnetism: an interdisciplinary ap- [244] K. Henrichsen, Encyclopedia of imaging science and proach, Vol. 66 (Springer Science & Business Media, technology (2002). 2013). 26

[263] S. J. Williamson and L. Kaufman, Journal of Magnetism [279] H. Shirzadfar, M. Nadi, D. Kourtiche, S. Yamada, and and Magnetic Materials 22, 129 (1981). T. Hauet, IRBM 36, 178 (2015). [264] G. L. Romani, Physica B+ C 126, 70 (1984). [280] K. Foster and J. Schepps, Journal [265] J. F. Schenck, Medical physics 23, 815 (1996). of Microwave Power 16, 107 (1981), [266] H. Koch, Journal of electrocardiology 37, 117 (2004). https://doi.org/10.1080/16070658.1981.11689230. [267] J. Vrba and S. Robinson, Superconductor Science and [281] D. Haemmerich, D. J. Schutt, A. W. Wright, J. G. Web- Technology 15, R51 (2002). ster, and D. M. Mahvi, Physiological measurement 30, [268] J. Mellinger, G. Schalk, C. Braun, H. Preissl, W. Rosen- 459 (2009), 19349647[pmid]. stiel, N. Birbaumer, and A. K¨ubler,Neuroimage 36, [282] E. Corsini, V. Acosta, N. Baddour, J. Higbie, B. Lester, 581 (2007). P. Licht, B. Patton, M. Prouty, and D. Budker, Journal [269] B. Riaz, C. Pfeiffer, and J. F. Schneiderman, Scientific of Applied Physics 109, 074701 (2011). reports 7, 1 (2017). [283] M. Szechy´nska-Hebda, M. Lewandowska, and [270] E. Boto, R. Bowtell, P. Kr¨uger, T. M. Fromhold, P. G. S. Karpi´nski,Frontiers in physiology 8, 684 (2017). Morris, S. S. Meyer, G. R. Barnes, and M. J. Brookes, [284] V. Jazbinsek, G. Thiel, W. MuEller,` G. WuEbbeler,` PloS one 11 (2016). and Z. Trontelj, European Biophysics Journal 29, 515 [271] J. Iivanainen, M. Stenroos, and L. Parkkonen, Neu- (2000). roImage 147, 542 (2017). [285] A. Fabricant, G. Z. Iwata, S. Scherzer, [272] J. Iivanainen, R. Zetter, M. Gr¨on,K. Hakkarainen, and L. Bougas, K. Rolfs, A. Jodko-W ladzi´nska, L. Parkkonen, Neuroimage 194, 244 (2019). J. Voigt, R. Hedrich, and D. Budker, [273] A. Borna, T. R. Carter, J. D. Goldberg, A. P. Colombo, bioRxiv (2020), 10.1101/2020.08.12.247924, Y.-Y. Jau, C. Berry, J. McKay, J. Stephen, M. Weisend, https://www.biorxiv.org/content/early/2020/08/13/2020.08.12.247924.full.pdf. and P. D. Schwindt, Physics in Medicine & Biology 62, [286] A. Yoshimi, K. Asahi, K. Sakai, M. Tsuda, K. Yogo, 8909 (2017). H. Ogawa, T. Suzuki, and M. Nagakura, Physics Let- [274] O. Alem, R. Mhaskar, R. Jim´enez-Mart´ınez,D. Sheng, ters A 304, 13 (2002). J. LeBlanc, L. Trahms, T. Sander, J. Kitching, and [287] P. Bevington, R. Gartman, and W. Chalupczak, Ap- S. Knappe, Optics express 25, 7849 (2017). plied Physics Letters 115, 173502 (2019). [275] E. Boto, N. Holmes, J. Leggett, G. Roberts, V. Shah, [288] M. Jiang, H. Li, Z. Zhu, X. Peng, and D. Budker, arXiv S. S. Meyer, L. D. Mu˜noz,K. J. Mullinger, T. M. Tier- preprint arXiv:1901.00970 (2019). ney, S. Bestmann, et al., Nature 555, 657 (2018). [289] C. E. Dreyer, A. Alkauskas, J. L. Lyons, A. Janotti, [276] R. Zhang, W. Xiao, Y. Ding, Y. Feng, X. Peng, L. Shen, and C. G. Van de Walle, Annual Review of Materials C. Sun, T. Wu, Y. Wu, Y. Yang, et al., Science Advances Research 48, 1 (2018). 6, eaba8792 (2020). [290] L. C. Bassett, A. Alkauskas, A. L. Exarhos, and K.- [277] R. Zhang, Y. Ding, Y. Yang, Z. Zheng, J. Chen, M. C. Fu, Nanophotonics 8, 1867 (2019). X. Peng, T. Wu, and H. Guo, Sensors 20, 4241 (2020). [291] M. Smiciklas and M. Romalis, “Test of Lorentz invari- [278] E. Elzenheimer, H. Laufs, T. Sander-Th¨ommes, and ance with Rb-21Ne comagnetometer at the south pole,” G. Schmidt, Current Directions in Biomedical Engineer- in CPT and Lorentz Symmetry (2014) pp. 95–98. ing 4, 363 (2018).